Skip to main content

Full text of "Journal of the Society of Motion Picture Engineers"

See other formats


From the collection of the 



d 

7. n 

X 






T>T 
o Jrrelinger 

u v Jjibrary 



t 



San Francisco, California 
2007 



JOURNAL OF THE SOCIETY OF 

MOTION PICTURE 
AND TELEVISION 

ENGINEERS 

THIS ISSUE IN TWO PARTS 
Part I, June 1951 Journal Part 77, Index to Vol. 56 



VOLUME 56 

January June 1951 



SOCIETY OF MOTION PICTURE 
AND TELEVISION ENGINEERS 

40 West 40th St., New York 18 



CONTENTS Journal 

Society of Motion Picture and Television Engineers 

Volume 56 : January June 1951 

Listen! below are only the papers and major reports from the six issues. See the 
Volume Index for those items which generally appear on the last few pages of each 
issue: Standards, Society announcements (awards, Board meetings, committee 
rosters and reports, Constitution and Bylaws, conventions, engineering activities, 
financial reports, membership, nominations, officer rosters and reports, section activ- 
ities), book reviews, current literature, new products and obituaries. 

January 

Current Problems in the Sensitometry of Color Materials and Proc- 
esses F. C. WILLIAMS 1 

A Direct-Reading Equivalent Densitometer A. F. THIELS 13 

A Versatile Densitometer for Color Films 

A. C. LAPSLEY and J. P. WEISS 23 
Recent Studies on Standardizing the Dubray-Howell Perforation for 

Universal Application W. F. KELLEY and W. V. WOLFE 30 

Effects of Television on the Motion Picture Theater 

B. SCHLANGER and W. A. HOFFBERG 39 
Some Comparative Factors of Picture Resolution in Television and 

Film Industries H. J. SCHLAFLY 44 

Image Tubes and Techniques in Television Film Camera Chains . . 

R. L. GARMAN and R. W. LEE 52 

Characteristics of All-Glass Television Picture Bulbs 

J. L. SHELDON 65 

Stereo-Television in Remote Control (Abstract) 

H. R. JOHNSTON, C. A. HERMANSON and H. L. HULL 75 

The Orthogam Amplifier (Abstract) 

C. L. TOWNSEND and E. D. GOODALE 76 
Diffuse and Collimated T-Numbers A Review and Description of 

New Equipment A. E. MURRAY 79 

The Differential Carbon-Feed System for Projection Arc Lamps . . 

A. J. HATCH 86 
Bibliography on High-Speed Photography 93 

ii Contents: Journal of the SMPTE Vol. 56 



February 

Image Gradation, Graininess and Sharpness in Television and Motion 
Picture Systems Part I: Image Structure and Transfer Charac- 
teristics OTTO H. SCHADE 137 

Technical Activities of the Motion Picture Research Council .... 

W. F. KELLEY and W. V. WOLFE 178 

Semiautomatic Color Analyzer LLOYD E. VARDEN 197 

Motion Picture Studio Lighting Committee Report. M. A. HANKINS 205 

A High-Speed Stereoscopic Schlieren System J. H. HETT 214 

Some Commercial Aspects of a New 16-Mm Intermediate Film Tele- 
vision System .... RAYMOND L. GARMAN and RLAIR FOULDS 219 
Television Film Recording and Editing .... ALBERT ABRAMSON 227 
The Genlock A New Tool for Better TV Programming (Abstract) . . 

JOHN H. ROE 232 

March 

High-Temperature Film Processing Its Effect on Quality 

RICHARD HODGSON and JACK HAMMER 261 
Ultrarapid Drying of Motion Picture Film by Means of Turbulent Air 

LEONHARD KATZ 264 
Television Transmission in Local Telephone Exchange Areas. . . . 

L. W. MORRISON 280 
A Professional Magnetic-Recording System for Use With 35-, 17J4- 

and 16-Mm Films 

G. R. CRANE, J. G. FRAYNE and E. W. TEMPLIN 295 
Carbon-Arc Characteristics That Determine Motion Picture Screen 

Light M. T. JONES and F. T. ROWDITCH 310 

The RCA PT-100 Theater Television Equipment 

RALPH V. LITTLE, JR. 317 

Projection Kinescope 7NP4 for Theater Television 

L. E. SWEDLUND and C. W. THIERFELDER 332 

Installation of Theater Television Equipment 

E. STANKO and C. Y. KEEN 343 

April 

Observer Reaction to Non-Simultaneous Presentation of Pictures and 

Associated Sound HAROLD N. CHRISTOPHER 369 

The Television Cameraman RUDY RRETZ 378 

A Simplified Index for Color of Illuminants 

FRANK F. CRANDELL, KARL FREUND and LARS MOEN 386 

A Rapid-Action Shutter With No Moving Parts 

HAROLD E. EDGERTON and CHARLES W. WYCKOFF 398 

Contents: Journal of the SMPTE Vol.56 iii 



A-C Magnetic Erase Heads . . . . M. RETTINGER 

A German Magnetic Sound Recording System in Motion Pictures . 

MARTIN ULNER 

How to Address a Professional Society KARL K. D ARROW 

High-Diffusion Screens for Process Projection (Abstract 

HUGH McG. Ross 
The Scientific Basis for Establishing the Brightness of Motion Picture 

Screens (A Discussion of Screen Brightness) 

FREDERICK J. KOLB, JR. 

May 

A Comprehensive Proposal for a Closed-Loop Theater Television 

System R. L. GARMAN and R. W. LEE 

Quality of Color Reproduction DAVID L. MACADAM 

A Time-Motion Study by Methods of High-Speed Cinematography . 
HENRY W. BAER, BERNARD F. COHLAN and ARTHUR R. GOLD 

16-M m Film Maintenance Cost and Analysis of Damages 

ERNEST TIEMANN and DENCIL RICH 

A New Theater Sound System . . . . B. PASSMAN and J. WARD 
The Cooling of Film and Slides in Projectors . . HUGH McG. Ross 

A 35-Mm Process Camera JOHN P. KIEL 

New All-Purpose Film Leader (A Report of the Subcommittee on 

Film Leaders) C. L. TOWNSEND 

Progress Committee Report CHARLES W. HANDLEY 



June 

Three-Dimensional Motion Picture Applications . . R. V. BERNIER 

Continuous Processing Machine for Wide Film 

HERBERT E. HEWSTON and CARLOS H. ELMER 
Slide Rule for Analyzing High-Speed Motion Picture Data .... 

KARL W. MAIER 

Use of Image Phototube as a High-Speed Camera Shutter 

ALSEDE W. HOGAN 

The New Visual Idiom NAT SOBEL 

Special Techniques in Magnetic Recording for Motion Picture Pro- 
duction GEORGE LEWIN 

Synchronous J-In. Magnetic Tape for Motion Picture Production . 

GEORGE LEWIN 

\r\\ Video Recording Camera . .F. N. GILLETTE and R. A. WHITE 
Practical Solution to the Screen Light Distribution Problem . . . 

CHARLES R. UNDERBILL, JR. 

iv Contents: Journal of the SMFFK Vol.56 



Current Problems in the Sensitometry 
of Color Materials and Processes 

By Franklin C. Williams 



The methods of seiisi tometry of color materials and processes are special- 
ized developments of the methods of black-and-white sensitometry. The 
nature of the individual material and its intended use govern the specifi- 
cations of the operations of exposing, processing, density measurement, 
and interpretation of results. Apparatus and techniques now available 
are adequate for important applications of sensitometry in the manufacture 
and use of color materials. Current research is refining existing methods 
of sensitometric investigation and yielding more significant test results. 



INTEREST IN THE SENSITOMETRY of 
color materials and processes has 
grown rapidly in the motion picture in- 
dustry, especially as the film user rather 
than the film manufacturer has become 
involved in processing, printing and 
other laboratory phases of color-film 
production. In laboratory operations 
involved in production of black-and- 
white motion pictures, sensitometric 
methods have found important uses. 
It is reasonable to expect that similar or 
even greater benefits from their use may 
be found in work with color motion 
pictures. Certainly, color-film manu- 
facturers have found color sensitometry 
useful to the point of necessity. Sensi- 
tometric methods of investigation and 



Communication No. 1361 from the Kodak 
Research Laboratories, a paper by Frank- 
lin C. Williams, Research Laboratory, 
Eastman Kodak Co., Kodak Park Works, 
Rochester 4, N.Y. It was read by Dr. 
W. T. Hanson, Jr., of Eastman Kodak 
Co. on October 11, 1949, at the Society's 
Convention in Hollywood, Calif. 



control have played an essential part in 
the development, production and proc- 
essing of the wide variety of sensitized 
color materials which are now available. 
The extent to which the usefulness of 
color sensitometry can soon be broad- 
ened is indicated by certain aspects of 
the present state of the science and by 
developments which are now in progress. 
It is the intent of this paper to present 
a brief view of this state, and of certain 
current developments, with respect to 
activities in the plants and laboratories 
of the Eastman Kodak Co. 

The general purpose of sensitometric 
investigation is the establishment of 
useful relationships among the elements 
of the chain of operations which result 
in a photographic image; or rather, 
finally, to relate the elements of image 
formation to the impression which the 
photographic image forms in the mind 
of the observer. Sensitometric tests 
usually are required to show how the 
quality of an image is dependent on ex- 
posure, on processing, or on the charac- 
teristics of the materials used. The re- 



January 1951 Journal of the SMPTE Vol. 56 



lationships among these and other ele- 
ments are described in terms of physical 
quantities. Sensitometry must, there- 
fore, provide accurate and practical 
means of making the required physical 
measurements. It must also select and 
specify what kind of measurement 
should be made in order that the re- 
sults be most significant. It must fur- 
ther provide systematic interpretation of 
the results. Not all of these things can 
be done at once. To date, development 
of the science of color sensitometry has 
emphasized specification of a few basic 
kinds of measurement, and development 
of accurate and dependable means of 
making them. Only during the past few 
years has vigorous development of sys- 
tematic interpretation of the results of 
these measurements been under way. 

In order that sensitometric testing be 
practical, a further requirement must be 
recognized: The test must be simple. 
It is for this reason that so much of the 
sensitometry of black-and-white ma- 
terials has been reduced to relationships 
between log exposure and image density. 
This relationship, usually expressed in 
the classic curve of Hurter and Drif- 
field, which is itself a model of sim- 
plicity, is derived from an irreducible 
minimum of straightforward opera- 
tions exposure, processing and deter- 
mination of image density. A sur- 
prisingly complete knowledge of the 
photographic properties of a film can be 
derived from such a test if it is prop- 
erly specified and conducted. It is 
natural that similar tests should be 
tried as the basis of the sensitometry of 
color materials and processes. Such 
trials have been adequately successful, 
so that most of the procedures of color 
sensitometry that are now in use are 
special adaptations of procedures orig- 
inally successful in the sensitometry of 
black-and-white materials. 

This adaptation of the procedures of 
the sensitometry of black-and-white 
materials has required throughout care- 
ful examination and usually extensive 



revisions of the specification of each 
element of sensitometric operation. 
Furthermore, particularly in specifying 
the objectives of density measurement, 
new concepts have had to be developed. 
A review of these revisions, develop- 
ments and some of the problems of color 
sensitometry can be made by examining 
the elements of routine testing pro- 
cedure in the order of their occurrence. 

Sensitometric Exposures 

Since exposure of the film is the first 
step of the routine, first met are prob- 
lems of knowing and of specifying 
what test exposures should be made, 
and how to make them. The images of 
color films are, of course, sensitively 
dependent upon the quality of the ex- 
posing light. In color-film testing, 
therefore, that quality must be ex- 
tremely well controlled. It must also 
be carefully chosen. In the film plane 
of a camera, a different quality of light 
exists at least at every differently 
colored point of the image. The re- 
sponse of the film to every one of these 
different qualities of light may be im- 
portant. No one quality of light, there- 
fore, can possibly furnish a complete 
test of the color-reproducing abilities 
of the film. An infinite set of qualities 
would be required. But in attempting 
to reduce tests to forms which are both 
simple and significant, effort is con- 
tinuously applied toward specifying, 
for routine tests, a small number of test 
colors which may test the film ade- 
quately. The exposing-light qualities 
which represent these colors are gen- 
erally chosen for one, or both, of two 
purposes. Some colors are used because 
they, and therefore their reproductions, 
are pictorially important. Tests using 
these colors give directly specific but 
limited information about the color- 
reproducing abilities of the film. Other 
exposures may not represent picture 
elements at all, but are chosen to give 
basic information about exposure-image 
relationships. From these relationships, 



2 



January 1951 Journal of the SMPTE Vol. 56 



the sensitometrist can determine in- 
directly certain general information 
about the color-reproducing abilities of 
the film. 

In sensitometric testing, both kinds 
of exposure are regularly used. Use of 
exposures which permit product ap- 
praisal by direct physical or psycho- 
physical measurements of color re- 
production has recently been consider- 
ably improved. Work by Brown, 
MacAdam, 1 and others is extending the 
data of color discrimination in rela- 
tionships involving both chromaticity 
and luminance. This is essential if we 
are to make quantitative comparisons 
of color reproductions under practical 
conditions, which generally involve ap- 
proximations in both chromaticity and 
luminance . In the Eastman Kodak Co. 's 
Color Control Dept., a research group 
working under R. M. Evans is engaged 
hi the study of the psychophysical and 
psychological factors involved in the 
perception of colored objects. Their 
findings are giving us a much better 
understanding of the influences of these 
factors in the evaluation of color re- 
production fidelity. Valuable data on 
the relative photographic importance of 
colors are coming from statistical analy- 
sis of the subject content of custom- 
ers' pictures. Although the results of 
these researches have as yet been ap- 
plied only to product-development work, 
they will soon find their way into routine 
testing procedures. 

Eventually, the reproduction criteria 
arising from such work will permit sys- 
tematic choice of the second kind of test 
colors those which may not be crit- 
ically important in themselves but 
which will provide sensitive and sig- 
nificant indicators of general reproduc- 
tion quality. With one exception, these 
colors cannot yet be chosen systemati- 
cally. The exception is the color gray. 
Although the intrinsic importance of ac- 
curate reproduction of gray is debat- 
able, a "gray-scale test" is a routine 
procedure in sensitometric testing of 



practically every kind of color film 
made. Years of testing have shown 
that it is a dependable, sensitive indi- 
cator of many important features of the 
color reproduction characteristics of a 
film. 

A sensitometric gray scale on a color 
film is the result of exposing the film to a 
series of intensities of white light. The 
gray-scale exposure, therefore, must be 
made with the kind of light which white 
or gray objects place on the film under 
conditions of normal use. More exact 
definition describes the white or gray 
objects as spectrally nonselective, dif- 
fusely reflecting objects. Specifying 
the conditions of "normal use" requires 
some investigation of the radiant energy 
source and of spectrally selective factors 
in the photographic system. 

For example, in exposures made with 
artificial light, the quality of light in 
gray-object images is primarily deter- 
mined by the original source, such as a 
tungsten lamp, but it is also importantly 
affected by spectral selectivity of the 
reflectors and lenses of the lighting units 
and by selective absorption in the 
camera lens. It cannot properly be 
simulated in the laboratory by simply 
matching the energy distribution of the 
unmodified lamp source. A special 
lamp-filter combination is required. 
Similarly, in exposures made with 
natural daylight, the white-light quality 
is dependent upon several factors. 
These include the position of the sun, 
the portion of sky effective as illuminant, 
the atmospheric conditions prevailing 
throughout the sky, the orientation of 
the subject with respect to the sky, the 
degree and nature of reflections by 
nearby objects, flare light, lens absorp- 
tions and other minor factors. The de- 
sign of sensitometer light sources must 
include consideration of these factors. 
We now have good approximations for 
some of the required light qualities; 
these include artificial daylight and 
sources duplicating tungsten lighting 
qualities. But the newer light sources 



Franklin C. Williams: Sensitometry of Color 



3 



in use in motion picture practice and 
elsewhere are difficult to match in a 
sensitometer, especially since sensitom- 
eter sources must be stable and ade- 
quately powerful as well as spectro- 
radiometrically correct. This problem 
is receiving considerable attention. For 
its solution, better spectroradiometric 
data are required than are now avail- 
able, and an instrument for this purpose 
is under construction in our laboratory. 
The actual exposure of film to the 
light quality selected can be made in 
sensitometers already developed for 
sensitometry of black-and-white ma- 
terials. The great importance in color 
sensitometry of making the exposure 
represent normal conditions of time and 
intensity, in order to avoid errors arising 
from failure of the reciprocity law, has 
led to development of special models of 
this apparatus, but no radically new 
principles have been introduced. 

Sensitometric Processing 

The second set of problems in color- 
film sensitometry is met in the process- 
ing of the sensitometric test films. It 
has long been a first principle of sensito- 
metric practice that the processing of 
test samples must satisfy two require- 
ments: It must be repeatable with ex- 
cellent precision, and it must be correct 
in kind. Probably no requirements of 
sensitometric color testing are more 
difficult to meet than these. Correct 
and invariable sensitometric color proc- 
essing is difficult, but where such proc- 
essing must be done, it is being done; 
frequently there is no really acceptable 
alternative course. Improved methods 
of using production processes for sensi- 
tometric tests are constantly under de- 
velopment, and improvement has been 
made, especially in treatment of the 
data, but variable processing can be 
used for evaluation of film character- 
istics only by making some sort of re- 
peated comparison with one or more 
selected "check" films, simultaneously 
processed to furnish a basis of reference. 



This familiar procedure offers a fre- 
quently useful substitute for direct 
measurement but also a somewhat 
treacherous one, since an adjustment 
based on one process-film combination 
often is not applicable to another. 
Modern methods of statistical analysis 
furnish means of recognizing such cases 
in a properly designed experiment, but 
not of correcting the data. 

The application of advanced statisti- 
cal methods to interpretation of sensito- 
metric data is a course of development 
in color sensitometry which deserves 
early recognition. Statistical methods 
are receiving wide recognition in indus- 
try as a tool especially useful in handling 
processes of complex variability. In 
color photography, known variables 
rarely can be made to operate with 
complete independence. It is possible, 
however, by special statistical methods, 
to extract from complexly variable data 
significant descriptions of the individual 
variations. The methods have been 
reduced to routine forms and are prov- 
ing of great value in both research and 
testing operations. 

A process which is repeatable but 
incorrect also presents difficulties. If 
the testing process does not accurately 
represent the normal treatment of the 
film, appraisal of the product is at 
least uncertain and is sometimes im- 
possible either by direct measurement 
or by comparison with a "check" film. 
Furthermore, considerable experimental 
evidence indicates that the best, and 
quite possibly the only, adequate 
method of maintaining long-time stabil- 
ity in characteristics of film and proc- 
essing, each independent of the other, 
is a combination of a stable reference 
process and extensive chemical analysis. 
Sensitometric color processing, there- 
fore, is the subject of much research 
activity. The research involves both 
special processing apparatus and special 
processing solutions. 

In sensitometry of materials for 
black-and-white photography, there has 



January 1951 Journal of the SMPTE Vol. 56 




Fig. 1. Automatic machine for processing sensitometric test strips 
of color films and papers. 



been adequate demonstration of the 
value of processing equipment especially 
designed for sensitometric testing. The 
reciprocating-paddle machine described 
by Jones, Russell and Beacham 2 has 
been notably successful in providing 
repeatable processing with excellent 
uniformity in its treatment of all 
samples in a particular loading. The 
principles of this machine have been 
applied to color processing with good 
success. Considerable enlargement of 
the equipment is, however, necessary 
to make color-processing output com- 
parable with the usual output of black- 
and-white film. For example, one kind 
of sensitometric color-processing ma- 
chine in use in the Film Testing Dept. 
at Kodak Park is shown in Fig. 1. 
These machines contain eleven de- 



veloping tanks and eleven wash tanks; 
the machines for developing black-and- 
white film contain only one of each. 
In the color-film processing machine, 
the test samples, on racks, are placed 
in light-tight developing units which 
are lowered by motor into the tanks. 
Each machine is equipped with three 
such developing units. Each unit has 
an independent set of paddles and an as- 
sociated drive. The unit of film, paddles 
and drive travels the length of the 
machine on an overhead track, then 
circles back to the starting end for re- 
use. Each unit carries with it a battery 
of 22 clock-driven controllers which 
automatically govern the sequence and 
duration of the process operations. The 
degree of agitation, electronically con- 
trolled, is widely variable and also is 



Franklin C. Williams: Sensitometry of Color 



automatically selected by the controlling 
panel. The three units, each holding 80 
sample strips and operating independ- 
ently, are spaced far enough apart in 
the machine to let the operator change 
processing solutions between units. It is 
therefore possible to process Koda- 
chrome, Ektachrome and Ektacolor 
films simultaneously on the same test- 
ing machine. 

Smaller machines with fewer auto- 
matic features are useful in process con- 
trol if they are designed to provide 
precisely controlled development. Sev- 
eral such machines have been built and 
have proved capable of producing proc- 
esses of good repetition accuracy. It is 
realized, however, that no sensitometric 
processing machine yet designed is 
ideal, particularly in its ability to imi- 
tate color processing as done in large 
continuous film-strip machines. Fur- 
ther improvements along this line are 
necessary. 

The processing machine is no more 
important than the solutions which go 
into it and is much more easily made 
free of undesirable variation. Sensito- 
metric processes must be repeatable 
not only from day to day, but, if neces- 
sary, from year to year. Such repeti- 
tion can be guaranteed only by identi- 
cal handling of the film in solutions of 
identical chemical constitution. A 
good processing machine will provide 
identical handling. A perfectly re- 
plenished continuous process would 
provide the identical processing solu- 
tions, but, although recent improve- 
ments in analytical methods promise 
to guarantee nearly perfect replenish- 
ments, this method has not been en- 
tirely satisfactory in sensitometric work. 
Instead, repeated identity of processing 
solutions has been accomplished by 
using, for each test, entirely new solu- 
tions mixed from a homogeneous re- 
serve stock of chemicals. 

The attainment of chemical identity 
in processing solutions is not a simple 
matter. Some solutions used in color 



processing are sensitive to extremely 
small changes in amount of ingredients 
and to variations in mixing procedure. 
So small a thing as the way in which a 
dry chemical is poured into the solution 
can materially affect the sensitometric 
result. We have found, for example, 
improved repetition precision when the 
alkaline ingredients of the solutions 
are dissolved in nearly the ultimate 
solution volume, before any oxidiz- 
able ingredients are present; other- 
wise, the amount of oxidation in the 
high pH region immediately surround- 
ing the dissolving alkali will vary ob- 
jectionably from batch to batch. Small- 
volume mixing requires precision weigh- 
ing, frequently with analytical balances, 
and liquid measurement with volumetric 
pipettes. If precision of chemical 
measurement and handling is adequate, 
repeatable mixes in one-gallon batches 
are possible in apparatus like that 
shown in Fig. 2, a jacketed cone with 
high-speed turbine homogenizer. It is 
preferable, however, to mix in larger 
batches wherever possible. Large- 
batch mixing is economically feasible 
only if the solutions can be stored, 
without deterioration, for later use. 
We have found that chemical solutions 
can be stored successfully by segre- 
gation of reactive components, and in 
difficult cases have found it helpful 
to store solutions just above their 
freezing point. Tests made through- 
out the past year have shown that 
even coupler-developer solutions can 
be made to maintain unchanged proper- 
ties for at least three weeks. 

When the film samples of a sensito- 
metric test have been exposed and 
processed, the testing routine requires 
next that their image densitities be 
determined. We recognize in color 
densitometry two classes of density 
measurement with two distinctly dif- 
ferent purposes: One class is called 
integral densitometry; the other, an- 
alytical densitometry. The purpose of 
integral densitometry is the measure- 



January 1951 Journal of the SMPTE Vol. 56 




Fig. 2. Jacketed conical mixing vessel and turbine homogenizer 
for mixing small batches of color-film processing solutions. 



ment of the composite, multilayer 
image to determine some total action, 
some particular effectiveness of the 
integrated image absorptions. The 
purpose of analytical densitometry is 
the determination of the individual 
densities of certain components of the 
image, such as the densities of the in- 
dividual yellow, magenta and cyan 
dyes. 

Printing Densities 

In the sensitometry of color mate- 
rials to date, the most valuable inte- 
gral densitometry has been the measure- 
ment of printing densities. A color 
negative, for example, may be printed 
on a color-print film that has a red- 
sensitive emulsion, a green-sensitive 
emulsion, and a blue-sensitive emulsion. 
The function of the negative is the 
regulation of the amount of exposure 
of each of these emulsions. The nega- 
tive performs this function by absorb- 
ing some of the light which would ex- 



pose these red-, green- and blue-sensi- 
tive emulsions. The amounts of the 
absorptions can be expressed as densi- 
ties called "printing densities." Since 
the negative image will usually absorb 
red, green and blue light unequally, it 
will have three different printing densi- 
ties. Each step of a gray scale will, 
therefore, have three printing densities, 
and, if these are plotted as a function 
of log exposure, the gray scale produces 
three characteristic curves. These are 
called the "gray-scale printing-density 
curves." A set of such curves is 
shown in Fig. 3. 

The determination of printing densi- 
ties by a densitometer requires the use 
of precisely specified kinds of red, green 
and blue light. The fundamental re- 
quirement is that the printing-density 
densitometer assign to radiant energies 
of various wavelengths the same rela- 
tive importances that would be as- 
signed by the printing system, that is, 
by the combination of printer light and 



Franklin C. Williams: Sensitometry of Color 



16 

I "' 2 

r 

= 08 



Log Exposure 

Fig. 3. Printing-density curves of 
the gray scale of a color-negative 
material. 

print material. In a simple case, if 
only light of X 680 m/i were effective in 
making a red-light print, the printing 
density of the negative would have to be 
determined by measurement with only 
light of X 680 m/i. In most practical 
cases, energy of a particular spectral 
distribution throughout a more or less 
broad band is effective in printing, and 
this distribution must be accurately 
reproduced in the density-measuring 



instrument. Since the densitometer 
must determine three printing densi- 
ties of each image, measuring energy of 
three such distributions must be readily 
available. In achieving these distribu- 
tions, we have thus far managed to 
obtain adequate accuracy by using 
optical filters of specially designed 
transmittance functions. Printing den- 
sities so determined have become ex- 
tremely valuable in product develop- 
ment and control. They are the prin- 
cipal sensitometric measurements used 
in performance inspection of color- 
negative materials. For this use, high- 
speed automatic densitometers have 
been developed. Figure 4 shows a recent 
model. Recently, the severe require- 
ments imposed by colored coupler- 
negative materials have made the de- 
sign of densitometer filters especially 
difficult and, as a result, research has 
been accelerated on the use of other 
means of spectrum selection. An ob- 
vious but difficult means is by disper- 




Fig. 4. Automatic recording densitometer for rapid determination 
of printing densities. 



January 1951 Journal of the SMPTK Vol. 56 



sion of the energy of the densitometer 
light source into a spectrum image, 
selection of part of this spectrum by a 
mask, recombination of the transmitted 
portion into a homogeneous mixture 
and use of this energy for the measure- 
ments. We have had an instrument of 
this type under development for several 
years. The difficulties of obtaining 
simultaneously the excellent spectral 
purity and considerable energy re- 
quired in the measuring beam are quite 
severe. 

Analytical Densities 

In some kinds of sensitometric work, 
image description by means of integral 
densities is inadequate. Especially in 
product development and in control of 
manufacturing and processing is knowl- 
edge of the individual dye densities of 
color images important. In such work, 
the desired characteristics of a color- 
film image can be expressed either by 
specifying its integral densities or by 
specifying the required densities of its 
component dyes. If specification is 
by integral densities, an obtained im- 
age can be compared with the desired 
image by integral densitometry, but, 
although this will describe the practical 
effects of any differences, it will not de- 
scribe the differences in a way that indi- 
cates where to apply corrective meas- 
ures. Wherever means are known by 
which individual dye deposits can be 
changed by known amounts in the proc- 
ess of image formation, approaches to 
the desired image characteristics are 
made most efficiently if the desired and 
obtained images are both described in 
terms of the individual dyes. Such a 
description is obtained by analytical 
densitometry and usually consists of a 
set of equivalent neutral densities. 
Equivalent neutral density is a unit for 
systematically expressing the densities 
of the individual dyes of a subtractive 
process in terms of their abilities to form 
grays. As defined by Evans, 3 the equiv- 
alent neutral density of a dye deposit 



in a subtractive color process is the 
density of the gray that would be 
formed by adding to that dye deposit the 
just-required amounts of the other 
dyes of the process. Figure 5 shows a 
set of equivalent neutral-density curves. 
Evans described an instrument for 
determining equivalent neutral den- 
sity, but other methods have since been 
developed for faster determination, less 
subject to an operator's judgment. 




Fig. 5. Equivalent neutral-density 
curves of the gray scale of a pro- 
fessional sheet-film material. 



One method, first described by Heymer 
and Sundhoff 4 in Germany, has been 
developed by the Film Testing Dept. in 
Kodak Park to a point that permits 
high-speed measurement. It requires 
an instrument that provides two identi- 
cal light beams; the sample, in one 
beam, is synthetically matched by a 
combination of dye wedges in the other 
beam. A recent model is shown in Fig. 
6. This model automatically makes all 
adjustments required in the analysis. 
The theory of analysis with instruments 
of this type involves fewer simplifying 
assumptions and approximations than 
are involved in other methods, but the 
cost of the instrument is high. Research 



Franklin C, Williams; Sensitometry of Color 



in the Kodak Laboratories has been 
directed toward the development of 
simpler means of analysis, particularly 
of one in which the required measure- 
ments are densities of the image deter- 
mined by use of three narrow bands of 
light, one each in the blue, green and red 
spectral regions. By a system of co- 
ordinate transformation, these three 
integral densities are made to yield the 
required equivalent neutral densities. 
Instruments capable of measuring the 
integral densities are made commercially 
by several manufacturers and have been 
made by many sensitometrists them- 
selves, but these instruments vary 
widely in their performance characteris- 
tics and abilities. Different instru- 
ments produce not only different inte- 
gral density values but also conflicting 



analytical density values. This condi- 
tion may be improved by activity of a 
committee of the American Standards 
Association which is attempting some 
standardization of color densitometry. 
In our laboratory, we use a densitometer 
of our own design and perform the co- 
ordinate transformations by a specially 
designed electrical analog computer, 
shown in Fig. 7. Integral density 
values are placed in it by setting three 
potentiometer dials. Closing one of the 
switches in the lower right-hand corner 
completes a circuit which instantly 
computes the equivalent neutral-den- 
sity value of the cyan dye and acti- 
vates a servo mechanism which ro- 
tates the central upper dial to the cor- 
rect equivalent neutral-density figure. 
Upon opening the first switch and clos- 




10 



Fig. 6. Automatic analytical color densitometer. 
January 1951 Journal of the SMPTE Vol. 56 



ing a second, the magenta equivalent 
neutral density replaces the cyan-den- 
sity figure; a third switch causes the 
yellow equivalent neutral density to 
appear. Accuracy of the computed 
transformed density value is better 
than 0.01. 

In this work, as elsewhere in color 
sensitometry, the elementary theory is 
simple, but improvements past the ele- 
mentary point involve labor among 
complex phenomena. A great deal of 
recent research on improvement of 
color-density measurements has been 
concerned with determining the real 
significance of image-analysis data. 
An essential step in this investigation 
is the derivation of an accurate spectro- 
photometric description of the mini- 
mum set of variables which can be 
combined to reproduce not only all the 
colors of which the process is capable 
but all the spectrophotometric distribu- 
tions as well. Mathematical procedures 
for handling this problem have been 
developed. The analytical compo- 
nents so determined are being used in a 
study of product and processing varia- 
tions to determine whether present 
methods of analysis yield the most sig- 
nificant data possible. 

The fact that this work is being done 
is evidence that development of color 
sensitometry even now is growing out 
of its first stage, in which emphasis has 
been placed on the improvement of 
precision and accuracy of measure- 
ment. It is passing into the next phase, 
in which measurements, made with 
techniques of adequate precision and 
accuracy already achieved, are to be 
applied to more significant tests. Prog- 
ress is being made toward the day 
when sensitometric methods may be 
as definitive in the specification of 
color-film quality as they are in the 
specification of quality of materials for 
black-and-white photography. That 
day has not yet come, but a great deal 
of progress has already been made. 

Present-day tests are valuable tests. 



They require precise application of a 
carefully chosen exposure, a correct, 
precisely controlled color process, and 
densitometry by an accurate, rapid 
instrument which measures a specific 
kind of density particularly suited to 
derivation of significant information. 
By use of these solidly founded elements, 
it is possible to draw sensitometric 
curves on which we can make signifi- 
cant, though still experimental, meas- 
urements of contrasts, gradients, speeds, 
densities, exposure latitude and other 
important features of the material. 
These things are being done hundreds of 
times every day, furnishing information 
which is reliable and definitive. 

Color sensitometry, therefore, stands 
now in a solid position of usefulness, 
with a good deal of accomplishment 
already behind it. Its immediate prob- 
lems are those of improvement and ex- 
ploitation of demonstrated techniques 
while pursuing a background develop- 
ment of new methods. The course of 
development will, in the near future as 
in the past, be considerably influenced 
by the demands of new products and 
new applications. 




Fig. 7. Electrical analog computer 
for determining equivalent neutral 
densities from narrow-band integral 
densities. 



Franklin C. Williams: Sensitometry of Color 



11 



References 

1. W. R. J. Brown and D. L. MacAdam, 
"Visual sensitivities to combined 
chromaticity and luminance differ- 
ences," /. Opt. Soc. Amer., vol. 39, pp. 
808-834, Oct. 1949. 

2. L. A. Jones, M. E. Russell and H. R. 
Beacham, "A developing machine for 
sensitometric work," Jour. SMPE, 
vol. 28, pp. 73-94, Jan. 1937. 

3. R. M. Evans, "A color densitometer for 
subtractive processes," Jour. SMPE, 
vol. 31, 194-201, Aug. 1938. 

4. G. Heymer and D. Sundhoff, "Uber die 
Messung der Gradation von Farben- 
filmen," Veroffentl. uriss. Zentral-Lab. 
phot. Abt. Agfa, vol. 5, pp. 62-76, 1937. 

Discussion 

DR. GUNDELFINGER (Chairman of the 
Session), to Dr. Hanson who delivered Mr. 
Williams's paper: Doctor, I might just 
point out one thing. I believe that densi- 



tometer wedges must consist, must they 
not, of the same components that are used 
in the color processes? 

DR. HANSON: Yes, they must be com- 
posed of those components that are used 
in the process. 

DR. CASPAR: Is the power- type agita- 
tion the generally adopted method? 

DR. HANSON: Yes, that is true. Both 
of the machines that are shown in the 
slides have paddle- type agitation. In both 
machines the paddles are variable in speed 
and in the larger machine, in pitch, so 
the distance from the surface of the film 
to the paddle may be varied. 

DR. CASPAR: Do they move parallel to 
the film? 

DR. HANSON: They move across a 35- 
mm film. I might add that the general 
type of machine that has been used has 
been described by Jones, Russell and 
Beacham in the Society's JOURNAL [ref. 2 
above]. 



12 



January 1951 Journal of the SMPTE Vol. 56 



A Direct-Reading 
Equivalent Densitometer 

By A. F. Thiels 



The definition of equivalent density of a primary color of a multilayer color 
film is given and a direct-reading photoelectric equivalent densitometer is 
described. The method of operation of the instrument is explained and 
the basic features of the electronic circuit and the optical and mechanical 
layouts are given. The apparatus has made it possible to make direct 
measurements of the density of any one primary color of a color film with- 
out being affected by the presence, if any, of other primaries. 



PRESENT-DAY COLOR FILMS Consist 
principally of three emulsion lay- 
ers in each of which, after exposure and 
color development, a color is formed: 
yellow in the blue-sensitive top layer, 
magenta in the green-sensitive middle 
layer and blue-green (cyan) in the red- 
sensitive bottom layer. These are 
called primary colors and will be re- 
ferred to as follows: 

j, yellow* 

m, magenta 

c, cyan (blue-green) 

In the subtractive color composition 
practically all color variations can be 
reproduced by varying the relative pro- 
portions of the color density of the pri- 

A contribution submitted March 22, 1950, 
by A. F. Thiels, Gevaert Photo-Products, 
Antwerp, Belgium. 

*Designation of yellow by j follows the 
practice of Bingham 4 in which footnote 2 
on p. 371 notes: The letters / and j are 
used instead of Y and y to represent den- 
sities in the yellow layer in order to avoid 
conflict with the notation of additive 
colorimetry. 



mary colors. In order to measure 
these color quantities in the sensi- 
tometry of color film, the equivalent 
density precept has come into current 
use. 1 - 2 

The equivalent density of a primary 
color is the neutral density which is ob- 
tained when the required quantities of 
two other primaries are added to the 
primary color to form a visually neu- 
tral gray. The determination of equiv- 
alent densities is therefore always linked 
to a selection of three primary colors, 
and is, furthermore, dependent on the 
lighting condition under which the film 
is visually examined. 

The significance of this precept be- 
comes more obvious when it is taken into 
consideration that a gray step wedge ex- 
posed in an intensity sensitometer 
should reproduce a wedge having all 
its steps neutral gray. The character- 
istic sensitometric curves of such a 
gray step wedge expressed in equiva- 
lent densities will by definition coincide 
(Fig. 1A). If, however, some of the 
steps are not neutral gray, the curves 
will no longer coincide; for example, 



January 1951 Journal of the SMPTE Vol. 56 



13 



o o o 




LOGAR. EXPOSURE 

Fig. 1A. Characteristic sensi- 
tometric curves of a well-bal- 
anced color step wedge. 



LOGAR. EXPOSURE 

Fig. IB. Characteristic sensi- 
tometric curves of an unbal- 
anced color step wedge. 




Fig. 2. Spectral dia- 
gram of a typical selec- 
tion of primary colors. 



500 WAVE LEM<3TH 60O 



700 



when they have a touch of blue, the 
"yellow equivalent curve" will be low- 
est. The amount by which the color 
balance has been disturbed can be read 
directly from the curves. 

For instance, on a step wedge of which 
sensitometric curves are as shown in 
Fig. IB, the lowest densities have a 
bluish hue, the medium ones are neu- 
tral and the highest densities will have 
a brownish tint because of the pre- 
dominance of yellow. 

For the purpose of determining the 



characteristic curve of each layer, it is 
not possible to separate the different 
layers of the material and measure- 
ments must be carried out on the multi- 
layer film as a whole. Different meth- 
ods can be used in order to arrive at 
more or less accurate evaluations of the 
sensitometric curves of the individual 
layers: (a) by the conversion of meas- 
urements at three different wave- 
lengths 3 ' 4 ; (b) by the recomposition of 
the color by means of three standardized 
primary color filters. 2 - 5 



14 



January 1951 Journal of the SMPTE Vol. 56 



PHOTO ELECTRIC CCLL 



MOVABLE 5CT 
OF PRISMS 




A.C. MAIMS 
125 V. 



Fig. 3. General layout of the equivalent densitometer. 



These methods are slow and, in addi- 
tion to elaborate calculations, require 
specially trained personnel. Hence, 
methods had to be evolved by which 
direct determination could be obtained 
of sensitometric characteristics of the 
complete multilayer color film. Such a 
method is afforded by the equivalent 
densitometer. 

The Equivalent Densitometer 

To facilitate the understanding of its 
operation, a study of the spectral dia- 
gram of a subtractive color layer will 
be of great assistance. Figure 2 shows 
the spectral-density curves of a typical 
selection of primary colors and also the 
density curve of the composition formed 
by them. The information which is 
sought is the proportion of j B of the 
yellow primary in the multilayer. 
Actually only the total density (through 
a narrow-cut blue filter) can be meas- 
ured: 

B = js + rns + CB 

If it were possible to subtract auto- 
matically the proportions of the second- 
ary absorptions, m B and C B , from the 
total measurement, the desired pur- 
pose would be attained. 

Assuming a linear relationship be- 



tween the secondary and the peak ab- 
sorption of a primary color (which ele- 
mentarily proves to be correct), thus, 
m B /m G = constant and C B /C R = con- 
stant with similar notations for other 
secondary absorptions, it may be in- 
ferred [see references 3 and 5] that the 
equivalent density, j, m and c of the 
three primaries, is found by the solu- 
tion of the linear system of three equa- 
tions: 

j = k u B - k lz G - kuR 

m = &21-B + k%iG kzsR (I) 

c = ky\B kszG H- kszR 

B, G and R are total measurements 
through narrow-cut blue, green and red 
filters. The constants, kn, . . . , 33, 
are positive numbers which are obtained 
from the absorption curves of the pri- 
mary colors. The equivalent densi- 
tometer automatically solves this prob- 
lem. 

General Scheme 

After passing through a heat-absorb- 
ing screen, the light of a stabilized 
underrated low-voltage lamp (Fig. 3) is 
collimated by a lens into a parallel 
beam which, by means of an adjustable 
set of totally reflecting prisms, is di- 
rected through one of the concentrically 



A. F. Thiels: Direct-Reading Densitometer 



15 



BALANCE 

SENSITIVITY \ CONTROL 

ZERO 




Fig. 4. Simplified layout of the electrical circuit of the equivalent 
densitometer. 



arranged filter sets situated in a revolv- 
ing disc. Through a second lens the 
rays are then converged and, after 
traversing a circular aperture and the 
sample strip to be measured, are di- 
rected onto the cathode of a photoelec- 
tric cell. 

When the prisms are adjusted to di- 
rect the rays through the outer set of 
filters situated in the revolving disc, 
the equivalent density of the yellow 
primary can be measured in this posi- 
tion. 

As the disc revolves at a constant 
speed of 2000 rpm, regular flashes of 
blue, green and red light strike the pho- 
toelectric cell and the currents gener- 
ated in the latter are amplified and con- 
ducted to the measuring instrument. 
By means of the polarity reverser 
(magnetic inductor) which is mounted 
on the shaft of the rotating disc, the 
blue flashes are caused to pass through 
the d-c meter in a positive direction, the 
red and green flashes, respectively, in a 
negative direction. Furthermore, since 
the lengths of the filter bands are so 
balanced that after amplification the 



light impulses are proportional with the 
constants, kn, &i 2 and &i 3 (shown in the 
first equation of the linear system I), 
it is possible to determine the quantity 
of j. Similarly, the other quantities, 
m and c, which are the equivalent 
densities of magenta and cyan, can be 
determined by adjusting the prisms to 
the corresponding filter segments of the 
rotating disc. 

At the outset, the "density" of the 
selective filters was such that the com- 
bination of light source, filter and pho- 
toelectric cell statically gave identical 
readings for the three filters. Later it 
will be seen that, in order to make cer- 
tain corrections, filter densities which 
are not always equal were adopted. 

Basic Circuit 

Figure 4 shows a simplified diagram of 
the amplifier circuit. The current gen- 
erated in the photoelectric cell by light 
impulses is amplified in the first triode 
tube, of which the upper part of the tube 
characteristic is used to obtain an al- 
most logarithmic amplification. The 
grid bias-resistance contributes toward 



16 



January 1951 Journal of the SMPTE Vol.56 



METER 
CURRENT 



RESULTAMT 
DIRECT 
CURRENT 




TIME 



|< - 1 REVOLUTlOfl OF FILTER DI5C 

Fig. 5. Diagram of periodic impulses through the meter. 



the achievement of this result. 6 The 
amplified signal is then transmitted to 
the parallel-coupled control grids of 
two heptodes. These act as a barrier 
and during one half revolution of the 
filter disc the left tube passes the signal, 
while during the other half revolution 
the right tube does likewise, in order that 
the direction of the current through 
the meter is alternatively reversed. 
This is attained by directing high 
square-wave voltages (40 v) which 
vary in phase by 180 degrees, to the 
modulation grids of the heptodes. 

The square-wave voltages are gen- 
erated in the magnetic inductor which 
is synchronized with the revolving filter 
disc.* A 7r-filter only lets pass the re- 
sultant direct current thus protecting 
the meter against excessive alternating 
currents and eliminating vibration of 
the needle. The meter is a d-c 100- 
microammeter of which the scale, 125 



* The same result can be obtained by pro- 
jecting light impulses of determined 
lengths synchronized with the revolving 
filter disc onto a set of photoelectric cells. 
Also, note the description of the improved 
electronic circuit given at the end of this 
paper. 



mm in length, is calibrated in equiva- 
lent densities. The scale reads from 
to 3. The shunt on the meter is so 
selected that a quick response of the 
needle is assured. 

It is possible to make an electrical 
circuit by which a logarithmic ampli- 
fication is obtained. Such an amplifica- 
tion gives a linear density scale over the 
whole measurable range. 6 Preference 
has been given to the adoption of a 
squared density scale. Although the 
intervals on this scale become somewhat 
short at the higher densities, they are 
nevertheless quite distinct and the 
accuracy in reading is not affected as 
is the case with logarithmic scales. 

This amplification has the advantage 
that by an adequate choice of the selec- 
tive filter densities certain apparent 
deviations of the Lambert-Beer law 
can be compensated, e.g., those caused 
by the curvature of the absorption 
curves of the selective filters, by fog 
other than that caused by dye com- 
ponents or by slight variations in the 
proportion of secondary and peak ab- 
sorptions of the primary colors in func- 
tion of density. 

In fact, a closer observation of the 
impulse registration (Fig. 5) reveals, for 



A. F. Thiels: Direct-Reading Densitometer 



17 



example, that the impulse through the 
meter, + k\\B = surface h X b, depends 
as much on the quantum of density 
(height h), which is the sum of the 
density of the (blue) selective filter and 
film density, as on the length of the 
filter (width 6). 

Changing the density of the filter and 
its length (while holding k n B constant) 
moves the operating range of the triode 
to a different portion of its nonlinear 
control characteristic. Thus different 
deviations from linearity could be ob- 
tained from the impulses provided by 
each filter. 

For a linear relation between density 
and meter current, the filter density 
may be varied providing the length of 
the filter is properly adjusted so that 
the surface remains b X h = knB.* 
This applies for whatever sample den- 
sity is placed in front of the photoelec- 
tric cell. When the scale is not linear 
but, for example, squared, deviations 
will occur. 

We shall not go further into these cor- 
rections now, but meanwhile it is clear 
that for constant-density disparities in 
the film strip, constant-current vari- 
ations will show on the meter when the 
relationship is linear, and furthermore, 
that these const antndensity disparities 
give no constant-current variations over 
several points of the meter scale, when 
the relationship ceases to be linear. 
These deviations allow compensation 
for the above-mentioned errors and 
they make the equivalent curves of the 
three primary colors coincide better. 

Calibration of the Apparatus 

The lengths of the filters in the circu- 
lar slits of the disc are adjustable by 
means of sliding cover plates. The 
length of dark spaces between filters 
has no influence on the result because the 
"barrier tubes" do not allow current 
to pass when no light strikes the photo- 
electric cell. Therefore, the dark sec- 



* (f.i. the broken-line rectangle) 



tions of the positive and the negative 
halves need not be equalized, which 
greatly simplifies the adjustment. 

For each filter segment the relative 
lengths of the positive and negative 
selective filters have to be so adjusted 
that the apparatus will perceive only one 
of the primary colors. When, for ex- 
ample, the light traverses the outer 
filter segment of the disc which measures 
the yellow equivalent density of a film 
strip, the meter should respond to a 
density variation of the yellow primary, 
but not for density variations of the two 
other primary colors. By a careful selec- 
tion of the filters and accurate adjustment 
of the filter lengths for any density of a 
primary color to be measured, needle 
deflections of less than 0.02 are obtained 
at any position on the scale for density 
variations of the other two primaries 
ranging from to 3, no matter whether 
these densities are placed together or 
separately in front of the photocell. 

After the ratios of lengths of the three 
filters have been established within each 
filter circle, these ratios must be pre- 
served during an additional adjust- 
ment. This adjustment consists of 
proportional changes in the over-all 
lengths of the filters hi the concentric cir- 
cles, so that the meter deviations should 
be identical for the three positions of the 
reflecting prisms when measuring a vis- 
ual neutral gray. This adjustment is 
necessary to make the densities of all 
three dyes register properly on a single 
meter scale. To this effect a series of 
visually neutral gray steps are carefully 
selected by a light of 3000 K. 

The absolute length of the filters is 
finally established when the shunt rheo- 
stat, which is necessary to adjust the 
meter to zero, just critically damps the 
meter. 

Lastly, the specular density of the 
neutral gray steps is determined with 
the aid of an optical densitometer 
(Martens' Polarisation Photometer) and 
on the basis of this measurement a scale 



18 



January 1951 Journal of the SMPTE Vol. 56 




Fig. 6. Practical construction of the instrument. 



is calibrated in specular equivalent 
densities.* 

The adjustment of the apparatus is 
quite simple. The prisms are first 
aligned with the middle filter segment 
(measurement of the equivalent ma- 
genta). A calibrated filter of density = 
3 is inserted in the head of the swivel 
measuring arm and placed in position in 
front of the photoelectric cell. By 
means of the regulator the needle of the 
meter is set on density 3 on the scale. 
The calibrating filter is then slid aside 
and the meter set to zero by a second 
control. The latter adjustment does 
not influence the former. The instru- 
ment is now ready for use. 

The apparatus is fitted with a ratchet- 
slide which allows the wedge to be ad- 
vanced layer by layer facing the aper- 
ture. This makes it possible to meas- 



* In this respect, it is not essential that the 
optical system should fill the requirements 
of specular measurements. 7 



ure the color wedge layer by layer and 
overcomes the necessity of readjust- 
ment of the prisms for each step. The 
ratchet-slide assures that at every move 
exactly the same area of the wedge 
faces the aperture. For routine work 
the measuring arm can be fixed just 
above the test wedge. 

The filter discs are made interchange- 
able so that the appropriate disc can be 
fitted for each set of primary colors. 
In theory it is possible to change the 
calibration of the apparatus electrically, 
but this implies the risk of errors and in- 
accuracies in the adjustment and there- 
fore preference is given to the inter- 
changeable discs. 

The photograph (Fig. 6) shows the 
practical construction of the instru- 
ment. 

Stability of the Apparatus 

Special care had to be taken with 
the stability of the apparatus in view 
of the industrial line fluctuations. 



A. F. Thiels: Direct-Reading Densitometer 



19 




20 



January 1951 Journal of the SMPTE Vol. 56 



With the apparatus under review, 
line voltage variations can be main- 
tained within == 1% by the introduc- 
tion of a magnetic voltage stabilizer on 
the input line. The built-in voltage- 
regulator tubes and current regulators 
assure a stabilization of less than 
0.1% for the crucial parts of the circuit. 
They completely check slow voltage 
changes over an interval of a few cycles. 
Short surges, if they occur at the mo- 
ment when the photoelectric cell receives 
an impulse, are absorbed by the heavy 
choke which protects the meter. In 
order to increase the stability, the fila- 
ment current of the first triode was re- 
duced to 200 ma and stabilized with a 
ballast tube. In this way, no readable 
changes in full-scale deflections are 
noticeable with line voltage variations 
from 75 to 140 v (nominal voltage being 
125v). Table I shows a series of meas- 
urements of the magenta primary of a 
gray step wedge for voltages ranging 
from 135 to 75 v. No additional adjust- 
ments were made during the measuring. 

Lacking a frequency generator, the 
systematic examination of the line fre- 
quencies is impossible. In practice, 
however, no effects of frequency vari- 
ations of the line are experienced be- 
tween 48.5 and 50.5 periods. 

The consumption of the apparatus is 
80 w and the warming-up time is about 
five minutes. 

The instrument has been in use in our 
laboratories for about nine months and 
proved to be reliable. A standardized 



neutral wedge was measured every three 
or four days over a period of approxi- 
mately two months. The maximum 
deviation recorded was 0.05, in the re- 
gion of density = 2. This deviation 
was partly due to inaccurate position- 
ing of the standard wedge in the ratchet- 
slide and may partly be attributed to 
the aging of the tubes and the photo- 
electric cell (variations in color sensi- 
tivity). 

Acknowledgment 

The author acknowledges the interest 
and advice of L. A. Meeussen, Gevaert 
Color-Film Dept., and F. T. Mees, radio- 
technician; and wishes to express appre- 
ciation to H. Verkinderen, Director of 
Research at the Gevaert Factories, Ant- 
werp, for permission to publish this 
paper. 

References 

1. G. Heymer and D. Sundhoff, "Ueber 
die Messung der Gradation von Far- 
benfilmen," Veroffen. Wiss. Zentral- 
Lab. phot. Abt. Agfa, vol. 5, pp. 62-76, 
1937. 

2. R. M. Evans, "A color densitometer 
for subtractive processes," Jour. 
SMPE, vol. 31, pp. 194-201, Aug. 
1938. 

3. G. Heymer, Film und Farbe, Max 
Hesseverlag, pp. 25-28, Berlin, 1943. 

4. R. H. Bingham, "Sensitometric evalu- 
ations of reversible color film," Jour. 
SMPE, vol. 46, pp. 368-378, May 
1946. 

5. N. Senger, Film und Farbe, Max Hesse- 
verlag, p. 13, Berlin, 1943. 



Table I. Influence of Line Voltage Variations on the Measurement of the 
Magenta Primary Dye of a Gray Step Wedge. 



Step No.: 1 



11 



13 



15 



17 



19 



135 v 


.20 


.31 


.43 


.67 


1.00 


1.30 


1.57 


1.86 


2.22 


2.53 


125 v 


.20 


.31 


.42 


.67 


.99 


1.30 


1.58 


1.87 


2.22 


2.53 


115v 


.20 


.30 


.41 


.68 


1.00 


1.30 


1.57 


1.86 


2.22 


2.51 


105 v 


.20 


.30 


.42 


.66 


.99 


1.29 


1.55 


1.85 


2.22 


2.51 


95 v 


.20 


.31 


.41 


.67 


1.00 


1.30 


1.56 


1.87 


2.22 


2.52 


85 v 


.20 


.31 


.42 


.66 


1.00 


1.31 


1.57 


1.87 


2.23 


2.53 


75 v 


.20 


.31 


.43 


.68 


1.00 


1.31 


1.57 


1.86 


2.23 


2.53 



A. F. Thiels: Direct- Reading Densitometer 



21 



6. M. H. Sweet, "A precision direct-read- 
ing densitometer," Jour. SMPE, vol. 
42, pp. 148-172, Feb. 1942. 

7. ASA Z38.2.5 (1946), "American stand- 
ard for diffuse transmission density," 
p. 8, American Standards Association, 
70 E. 45th St., New York 17. 

[ADDENDUM: Since this paper was 
submitted, the author has developed an 
improved circuit which is reported to have 
given entirely reliable service during the 
last half of 1950. The author has kindly 
supplied the diagram and brief description 
for inclusion at press time. Ed. ] 

Improved Electronic Circuit 

It is possible to use photoelectric cells 
to generate the synchronized square 



wave. As the construction with photo- 
electric cells, to replace the magnetic 
inductor, is of more universal practice, 
we here describe a complete circuit 
(Fig. 7) showing the disposition of the 
cells. 

The use of photocells makes it possi- 
ble, in addition, to take the length of the 
"positive filter" longer than 180 (e.g., 
"positive filter" 240 "negative filter" 
120) so that the circumference may be 
more advantageously utilized. 

The addition of an amplifier stage be- 
hind the barrier-lamps has the advan- 
tage that a more robust meter can be 
used (1- to 3-ma), whilst the amplifying 
characteristic of this stage can be so 
selected that a linear-density scale can 
be drawn on the dial. 



22 



January 1951 Journal of the SMPTE Vol. 56 



A Versatile Densitometer for Color Films 



By A. C. Lapsley and J. P. Weiss 



A new densitometer for the analysis of color films reads densities with a 
narrow wavelength band at any desired wavelength between 350 and 760 
m/u. This instrument was constructed utilizing two commercially avail- 
able units : a Golem an Model 10- S Double M onochromator Spectrophotom- 
eter and a Western Electric RA-1100-B Densitometer. It has performed 
quite satisfactorily during more than two years of continuous service. 



RESEARCH STUDIES of Color film 16- 
quired an instrument for measur- 
ing the spectral densities of dye images. 
For maximum utility, the instrument 
had to meet a number of specifications. 
First, density measurements were to be 
made with essentially monochromatic 
illumination. Second, provisions for 
measuring density at any wavelength 
were desired, because for research pur- 
poses density readings made at the 
wavelength of maximum absorption of a 
given dye were most useful. Another 
requirement was the ability to read to 
quite high densities, at least 4.0. This 
requirement was even more important 
for color densitometry than for black- 
and-white, since the spectral density of 
a dye may exceed appreciably the neu- 
tral density to which it contributes. 
Accuracy was another obvious require- 
ment. To measure subtractive dyes ac- 
curately at high densities, only a very 



Presented on October 17, 1950, at the 
Society's Convention at Lake Placid, 
N.Y., by A. C. Lapsley and J. P. Weiss, 
Technical Div., Photo Products Dept., 
E. I. du Pont de Nemours & Co., Inc., 
Parlin, N. J. 



low percentage of stray white light could 
be tolerated in the monochromatic 
beam. Rapid and convenient operation 
was also specified. The instrument had 
to be operated by nontechnical person- 
nel with sufficient rapidity to handle a 
large volume of color-film sensitometric 
strips. Reliable, trouble-free operation 
was also highly important. 

Since none of the commercially avail- 
able color-measuring instruments com- 
bined all the desired features, it was 
necessary to design one. In the inter- 
ests of low design cost and maximum 
reliability an effort was made to utilize 
existing, proven components wherever 
possible. 

Description 

A special densitometer was con- 
structed, incorporating a modified Cole- 
man Model 10-S DM Spectrophotom- 
eter (made by American Instrument 
Co.) as the light-source unit and a 
Western Electric RA-1100-B Densitom- 
eter as the indicator. To obtain suf- 
ficient sensitivity to the radiant energy 
transmitted by the colored images, it 
was necessary to use a multiplier photo- 



January 1951 Journal of the SMPTE Vol. 56 



23 



GOLEMAN DOUBLE MONOCHROMATOR 
[__ OPTICAL SYSTEM 




Fig. 1. Schematic layout of Color Densitometer. 

Light originating at source S is resolved into monochromatic beam by the 
diffraction gratings, GI and G2. Exit beam is focused by lens Li4 onto den- 
sity located at A. Transmitted light is received by multiplier phototube M 
which is in turn electrically connected to the Western Electric Densitometer. 



tube. Direct-current voltage for the 
phototube was provided by a rectifier 
and filter operating from the a-c lines. 

The schematic diagram of the color 
densitometer appears as Fig. 1. The 
portion enclosed within the dotted lines 
is the original Coleman optical system. 
The light source, S, is a lamp with a 
vertical coiled, line filament which acts 
as the entrance slit of the monochroma- 
tor. The first transmission grating, GI, 
cemented to condenser lens LI, forms 
its spectrum across a fixed slit, Si, 
which passes a narrow spectral band 
into the second dispersing system. 
This light is collimated by lens L 2 and 
is reflected by a right-angle prism, P, 
through the second grating, G 2 , and is 
focused on slit 82 by lens L 3 . The de- 
sired wavelength is selected by rotating 
the cam, W, which, linked by arms, 
swings source-slit S and rotates prism P 



so that the spectrum is swept across 
slit S 2 , and the same wavelengths pass 
through both Si and S 2 at all times. 

Added to the Coleman Spectro- 
photometer is a filter system, F, to re- 
duce residual stray white light to a neg- 
ligible amount. While the double grat- 
ing monochromator passes only a small 
amount of stray white light, stated as 
being a fraction of a per cent, the re- 
quirement of light purity is very strin- 
gent if dye densities up to 4.0 (a transmit- 
tance of 0.01%) are to be read. The 
subtractive dyes used in color photogra- 
phy have fairly narrow absorption 
bands, roughly one-third the visible 
spectrum, and they transmit the other 
two-thirds of the spectrum quite freely 
To minimize errors from this cause, the 
appropriate one of five fairly narrow 
band-pass filters may be put into the 
light beam. The transmittance of these 



24 



January 1951 Journal of the SMPTE Vol f 56 



100 



90 - 



70 



60- 



50 



40 



30 - 



20 - 



10 




400 



500 600 

WAVELENGTH MILLIMICRONS 



700 



Fig. 2. Transmission curves of auxiliary filter system. 

Each filter combination transmits only a limited portion of the 
spectrum and eliminates the major portion of any stray light that 
leaks through the monochromator system. 



filters is shown in Fig. 2, where it is 
seen that they give a fairly complete 
coverage of wavelengths in the 400- to 
700-m/z range. 

To provide a pulsating signal to the 
Western Electric amplifier, which is 
tuned to a frequency of 450 cycles/sec, 
the light is interrupted by a chopper, 
C. This consists of a 15-slot disc 
driven by an 1800-rpm synchronous 
motor, just as in the original light 



source of the Western Electric Den- 
sit ometer. 

Additional elements to complete the 
light-source unit are a lens, L 4 , to focus 
an image of exit slit, S 2 , at the film aper- 
ture, A, and two mirrors which cause 
the final image to be oriented properly. 
The film aperture and the sensitometric 
strip-holder are identical with those on 
the Western Electric instrument. The 
multiplier phototube, M, mounted be- 



Lapsley and Weiss: Densitometer for Color 



25 



low the aperture, A, receives the light 
transmitted by the film. The output 
of the phototube is electrically con- 
nected to the amplifier of the Western 
Electric Densitometer. 

Figure 3 is a photograph of the light 
source unit with housing removed. 
The L-shaped casting contains the 
Coleman optical elements. The light 
source is at the left of the casting. The 
knob and dial near the light source 
comprise the wavelength control. The 
knob at the right of the casting is for 
the filter system. Clearly seen is the 
slotted interrupter disc and its motor. 
The black housing in front of the disc 
contains the lens and mirror system 



which focuses the exit slit of the mono- 
chromator on the photographic film 
being analyzed. 

The light transmitted by the film 
sample falls on the multiplier photo- 
tube mounted below the film aperture. 
A 1P22 tube was selected as having 
appropriate spectral sensitivity. It 
has considerable sensitivity at 700 m/i, 
where response is wanted, but falls off 
rapidly hi sensitivity at about 750 mju. 
Infrared response must be kept low 
since the infrared transmission of most 
organic dyes might otherwise lead to 
spurious density readings. 

The wiring diagram is shown in Fig. 
4. Direct-current voltage for the multi- 




Fig. 3. View of light-source unit with housing removed. 

Large casting to the rear contains Coleman double monochromator optics. The 
motor-slotted disc combination at the exit of the housing serves as a light interrupter 
which allows an a-c electrical signal to be picked up from the phototube. Housing on 
the near side of the motor contains a mirror-lens system for focusing and orienting the 
light from the exit slit onto the density to be analyzed. 



January 1951 Journal of the SM PTE Vol/56 




IIOAC 



Fig. 4. Wiring diagram of the Color Densitometer. 

High voltage supply to the multiplier phototube, Vi, and low voltage supply 
to the light interrupter motor, MI, and light source, Si, are shown. 




Fig. 5. The mounting of Color and Western Electric Densitometers. 

The meter on the Western Electric unit has been rotated 
to face the Color Densitometer operator. 

Lapsley and Weiss: Densitometer for Color 



27 



plier phototube dynodes is provided by 
rectifying and filtering the output of a 
high-voltage transformer. The two- 
section filter reduces the residual a-c 
signal to a level too low to cause errors, 
even at high densities. Since the out- 
put of the 1P22 as a function of wave- 
length is quite nonuniform, a gain con- 
trol independent of that available in 
the Western Electric amplifier is needed. 
This is supplied by the autotransformer, 
Ti, which varies the supply voltage to 
the 1P22. By doing this rather than 
varying the intensity of the light source, 
the maximum signal-to-noise ratio is 
maintained at any wavelength setting. 
Line-voltage fluctuations are compen- 
sated by using a voltage stabilizer. 
The other wiring shown in Fig. 4 is 
for the light source and synchronous 
motor. 

The signal output from the 1P22 
phototube is fed directly to the first 
amplification stage of the Western 
Electric RA-1100-B Densitometer,* 
an instrument very well known in the 
motion picture industry. The con- 
nection is made, through a shielded co- 
axial cable, in parallel with the No. 929 
phototube of the Western Electric 
Densitometer; this allows use of the 
instrument as either a black-and-white 
or a color densitometer without switch- 
ing. 

The complete color densitometer as- 
sembly is shown in Fig. 5. The mono- 
chromatic light source is to the right 
and the Western Electric Densitometer 
to the left. It may be noted that the 
density meter of the latter has been ro- 
tated to face the operator of the color 
densitometer. The meter has been 
mounted on a column and may be 
swung to face an operator of either 
instrument. 



*J. G. Frayne and G. R. Crane, "A preci- 
sion integrating sphere densitometer," 
Jour. SMPE, vol. 35, pp. 184-200, Aug. 
1940. 



Performance 

As noted in the introduction of this 
paper, one of the requirements of the 
color densitometer was a high degree 
of accuracy. This has been checked, 
both for phototube and amplifier linear- 
ity, and for the presence of minute 
amounts of stray white light in the 
monochromatic beam. Linearity tests 
were made with specially constructed 
neutral density "filters." These were 
thin brass discs perforated with a 
series of uniformly spaced holes. Their 
densities were calculated from the size 
and spacing of the holes and checked 
experimentally on an accurate photom- 
eter. Two such discs, having densities 
of 0.495 and 1.015, were on hand. 
At various selected wavelengths, these 
filters were individually introduced 
into the light beam and the indicated 
densities recorded. Then with the 
filters removed, the light intensity was 
reduced until the meter indicated the 
value recorded for the 1.015 filter. 
At this light level the two discs were 
again introduced into the beam and the 
densities recorded. This process was 
repeated until an indicated top density 
of 4.060 was reached. The results are 
shown in Table I. 

Up to the top density of 4.00 it is 
seen that there is good linearity at wave- 
lengths throughout the visible spectrum. 
The shouldering that appears at 350 
and 760 mju at high densities is probably 
the result of phototube noise caused by 
the high voltage that has to be applied 
to it at these wavelengths. 

The above checks, however, would 
not indicate the presence of stray white 
light. This can be serious, for if there 
is 0.01% unwanted light which is not 
absorbed by the selective dye, it will 
cause a density error of 0.002 at a den- 
sity level of 2.0, 0.02 at a level of 3.0, 
0.12 at 3.5 and 0.32 at 4.0. This rapid 
increase of the error at high densities 
suggests a ready method of checking. 
This is to measure the density of two 
selective absorbers at the level of 2.0 



28 



January 1951 Journal of the SM PTE Vol. 56 



Table I. Densities measured 
at various wavelengths. 



Table II. Density checking results. 



True 
Den- 
sity 


Wavelengths 


',i 


m/i 






350 


440 


540 


700 


760 


0.495 
1.015 
1.510 
2.030 
2.525 
3.045 
3.540 
4.060 


0.51 
1.03 
1.51 
2.05 
2.54 
3.03 
3.40 




1 
1 
2 
2 
3 
3 
4 


.50 
.02 
.52 
.04 
.51 
.06 
.55 
.02 




1 
1 
2 
2 
3 
3 
4 


.495 
.025 
.52 
.04 
.54 
.065 
.55 
.04 




1 
1 
2 
2 
3 
3 
4 


.50 
.02 
.52 
.04 
.54 
.06 
.57 
.01 


0. 
1 
1, 

2 
2 
3 
3 


505 
035 
,54 
06 
55 
.05 
.42 



and to measure the density of their 
combination. Such a test has been car- 
ried out on a sample of color film in the 
wavelength region 600-700 m/x, with the 
results tabulated in Table II. It is 
seen that there is no noticeable error 
caused by stray light. 

The color densitometer has given 
quite trouble-free performance. It has 
been in daily use for over two years. 
The only attention required has been 
occasional replacement of the lamp. 

Discussion 

M. C. TOWNSLEY: Is the receiver which 
receives the energy after it passes through 
the film so arranged that it reads diffuse 
density or is it substantially specular 
density? 

MR. LAPSLEY: Actually, it probably 
reads a combination, but closer to specular 
density. That is, we don't have any in- 
tegrating sphere. We use light which is 
focused onto film and which diverges be- 
yond the film. The phototube is mounted 
close enough so that it catches all of the 
light that passes through the film, or at 



Wave- 
length, 


Strip 1 Strip 2 


Strips 
Calc. 


1 & 2 
Read 


600 


1 


.30 


.18 


2.48 


2.49 


610 


1 


.33 


.21 


2.54 


2.54 


620 


1 


.39 


.27 


2.66 


2.66 


630 


1 


.52 


.37 


2.89 


2.90 


640 


1 


.61 


.44 


3.05 


3.03 


650 


1 


.68 


L.51 


3.19 


3.18 


660 


1.78 


L.58 


3.36 


3.34 


670 


1 


.82 1.63 


3.45 


3.44 


680 


1 


.87 1.66 


3.53 


3.52 


690 


1 


.88 1.67 


3.55 


3.56 


700 


1 


.86 ] 


L.65 


3.51 


3.50 



least a major portion of it. This instru- 
ment as we built it was designed primarily 
for color-film work, using dyes which 
have only a negligible amount of scatter- 
ing, and it is our opinion, which has been 
checked up to the limits that we can check, 
that the density it measures would ac- 
tually be the specular and diffuse density. 

MB. TOWNSLEY: Do you feel that, for a 
dye material, the specular density and 
diffuse density are not very different? 

MB. LAPSLEY: That is correct. We of 
course could not make measurements on 
black-and-white film with that instru- 
ment, but there would be no point in doing 
so. 

ANONYMOUS: Have you found any dif- 
ficulty with selective fatiguing of the mul- 
tiplier phototubes? 

MB. LAPSLEY: We have not found any 
difficulty with that as such. Mainte- 
nance of a constant relationship of out- 
put versus wavelength is not required for 
proper operation of this instrument. 
Zero adjustment is convenient and is 
made for the wavelength selected just 
before density measurements are made. 



Lapsley and Weiss: Densitometer for Color 



29 



Recent Studies on Standardizing 
the Dubray-Howell Perforation 
for Universal Application 

By W. F. Kelley and W. V. Wolfe 



The adoption of safety base film throughout the motion picture industry 
has required the abandonment of the Bell & Howell perforation for color 
release prints. This fact presents an opportunity to achieve the long- 
desired goal of a single standard perforation for negative and positive films 
in all applications. Tests are described and conclusions reached covering 
registration problems in the studio, studio laboratory and release labora- 
tory, as well as accelerated and normal release life tests on Dubray-Howell 
perforated black-and-white prints. 



rriHE IMPORTANCE of the perforations 
J_ on the side of a motion picture film 
would be difficult to overstate. Those 
perforations are relied upon for propul- 
sion and registration in every photo- 
graphic and projection operation in the 
making and exhibiting of a motion pic- 
ture. Unfortunately, the importance 
of these perforations is not understood 
by a great many people in the industry, 
and even those who do realize their im- 
portance are often inclined toward the 
philosophy that "what was good enough 
for my father is good enough for me." 

The history of the perforation size and 
shape is contained in the JOURNALS of 
this Society and that information was 
very excellently gathered and presented 
by the Film Dimensions Committee in 



Presented on October 19, 1950, at the 
Society's Convention at Lake Placid, 
N.Y., by W. F. Kelley and W. V. Wolfe, 
Motion Picture Research Council, Inc., 
1421 N. Western Ave., Hollywood 27, Calif. 



the April, 1949, JOURNAL, at which time 
it was proposed, for the third time, that 
the Dubray-Howell perforation should 
be adopted as a universal standard. 

Just to review this situation briefly, 
note that the first accepted standard per- 
foration was the familiar Bell & Howell 
perforation which, prior to 1923, was 
standard throughout the industry for 
both negative and positive use. Be- 
cause of nonstandard projector sprock- 
ets, the inherently weak tear-resistance 
of the Bell & Howell sprocket was aggra- 
vated. A number of the pioneer engi- 
neers of the industry and of this Society 
considered the problem and came up 
with the present Eastman positive per- 
foration which was accepted throughout 
the industry for release print purposes. 
Even at that time, however, there were 
many voices raised in opposition to a 
different standard perforation for nega- 
tive and positive applications. 

In 1932 Messrs. Dubray and Howell 
proposed a perforation combining the 



30 



January 1951 Journal of the SMPTE Vol. 56 



rectangular shape of the positive per- 
foration with the 0.073-in. height of the 
negative perforation, thus obtaining the 
best features of both perforations from 
the standpoint of existing equipment, 
registration and projection life. Never- 
theless, in 1933 this Society adopted the 
Eastman positive perforation as the 
universal standard for both negative 
and positive film. However, the indus- 
try refused to accept this universal 
standard because it required changing 
every camera, projector or printer 
throughout the world. 

In 1937 the Subcommittee on Film 
Perforation Standards recommended 
that the 1933 standard be withdrawn 
and again proposed the Dubray-Howell 
perforation as the universal standard. 
This report was turned down by the 
Standards Committee because it was 
felt at that time that the large amount 
of background film accumulated in the 
libraries would prevent the universal 
perforation from being used. 

Beginning in 1947 and continuing 
since that time, your Film Dimensions 
Committee, under the chairmanship of 
Dr. E. K. Carver, has continuously 
had on its agenda the problem of secur- 
ing a universal standard perforation 
acceptable to all of the industry. Tests 
made by many people predominently 
supported the early contention of 
Dubray and Howell that this rectangu- 
lar perforation with an 0.073-in. height 
was satisfactory for all negative and 
positive purposes. M. G. Townsley at 
Bell & Howell demonstrated in some 
tests made not long ago that Dubray- 
Howell perforated film would operate 
with satisfactory steadiness in a camera 
equipped with a Bell & Howell full- 
fitting pilot pin. 

This situation might have continued 
without any conclusion for a long time 
but for the introduction by Eastman 
Kodak Co. of the new safety base film. 
Prior to this time, all of the commer- 
cially used color systems employed 
Bell & Howell perforated release prints 



because of the need for a high degree of 
registration in making such prints; but, 
because the new film base is reported 
to be somewhat lower in its tear strength 
than the nitrate film base, two color sys- 
tems adopted the Dubray-Howell per- 
foration and are currently using it. 
Both Trucolor and Cinecolor in making 
this decision found that they could suc- 
cessfully register from Bell & Howell 
perforated negatives to Dubray-Howell 
perforated color prints. 

Technicolor, unfortunately, although 
fully aware of the industry's long struggle 
for a universal perforation and of the suc- 
cessful use of the Dubray-Howell perfor- 
ation by other color companies, adopted 
the Eastman positive perforation with- 
out consulting or advising the producing 
companies of that decision. Perhaps 
Technicolor did not realize the studio 
significance of this decision. However, 
when studio photographic effects de- 
partments were notified that after a 
certain time all Technicolor prints 
would be supplied with Eastman posi- 
tive perforations, it became immediately 
evident that process projectors and per- 
haps other studio-owned precision equip- 
ment would require interchangeable 
movements in order to handle both 
Technicolor prints and black-and-white, 
or prints of any other color system. 
The cost of duplicating such movements 
is in itself moderately high, but what is 
much more important, such a situation 
materially adds to the danger of con- 
fusion and delay in any operation in- 
volving process projection photography. 

The matter was called to the atten- 
tion of the Research Council and a meet- 
ing involving all those interested, from 
manufacturers, commercial laboratories 
and studios, was held. As a result of 
this meeting, a comprehensive series of 
tests was laid out by the Research Coun- 
cil in the hope that the industry could 
be convinced that this was the tune to 
adopt a universal standard perforation 
and thus for all tune avoid any further 
confusion and expense which must in- 



Kelley and Wolfe: Dubray-Howell Perforation 



31 



STD.POS. PERR ST'O. NEGPERR DUBRAY HOWELL PERF. COOKE PERF 



PROJECTRJ 

TYPE Y//////, 

HJfc 






Fig. 1. Drawing showing the fit of various perforations, 
pilot pins and sprocket teeth. 



evitably result from what used to be a 
double standard and is now a triple 
standard. 

Four perforations were considered: 
the Bell & Howell, Eastman positive, 
Dubray-Howell and Cooke (Fig. 1). 
Experience and history had already 
eliminated the Bell & Howell and the 
Eastman positive perforation as candi- 
dates. Discussion with experts in print- 
ing problems, particularly having to 
do with the continuous type of printer 
on which better than 90% of all release 
prints are made, revealed that the 
square end of the Dubray-Howell per- 
foration was preferred over the rounded 
end of the Cooke perforation. It was 
also the belief of many industry experts 
that in other problems of registration, 
the Dubray-Howell perforation was 
superior to the Cooke. As a result, 
efforts were confined entirely to the 
Dubray-Howell perforation. 

Generally speaking, there are two 
problems involved: one is projection 
life and the other is registration. Each, 
of course, has a variety of important 
problems under that general heading. 
Study of projection life was divided into 
two parts : accelerated tests and normal 



release tests. Actually, there is already 
considerable experience in normal re- 
lease through the color systems which 
are using the Dubray-Howell perforation 
commercially, but it was recognized that 
these color prints present a different 
projection-life problem than normal 
black-and-white prints. Accelerated 
life tests on black-and-white prints, 
made by other investigators on care- 
fully aligned projection equipment, have 
shown approximately 10% greater life 
with the Eastman positive perforation 
than that obtained with the Dubray- 
Howell perforation. It was, however, 
recognized by the engineers in charge of 
these tests that normal release condi- 
tions might indicate a different answer 
because theater projectors are not uni- 
versally as well aligned as these test pro- 
jectors. Accordingly, the accelerated 
test was deliberately made in a projec- 
tor out of alignment and using badly 
worn sprockets. Figure 2 is a photo- 
micrograph of one of these sprocket 
teeth. Since it is badly undercut on 
both sides of the tooth, it has probably 
been reversed in the machine at some 
time during its life. This test reel was 
run approximately 300 times before the 



32 



January 1951 Journal of the SMPTE Vol. 56 






test was stopped. Although the film 
was not run to destruction, it was evi- 
dent at this time that in a machine as 
badly aligned as this one, the film could 
not run many more times. 

Figure 3 shows a photomicrograph 
of one corner of the Eastman positive 
perforation at the end of the running, 
and Fig. 4 shows a similar corner of a 
Dubray-Howell perforation. In both 
cases the tooth interfered at the corner 
of the perforation and caused a serious 
rupture of the film. Inspection of about 
80 ft of each of the two prints involved 
in the reel led to the conclusion that the 
Dubray-Howell perforation was stand- 
ing up a little bit better under this par- 
ticular test than the Eastman positive 
perforation. While this is contrary to 
the projection-life tests previously re- 
ferred to, it is not an unexpected differ- 
ence, because the smaller radius of the 
corner fillet in the Dubray-Howell per- 
foration means that the straight portion 
of the perforation is longer than is the 
case in the Eastman positive perfora- 
tion; thus, corner interference will be- 
gin with an Eastman perforation before 
it begins with a Dubray-Howell per- 
foration. 



As this article is being written, the 
partial release test has not yet been 
completed, but there is a picture in re- 
lease in the Los Angeles exchange area in 
which half of the 1000-ft release is per- 
forated with the Eastman positive hole, 
and the other half uses the Dubray- 
Howell perforation. These are so stag- 
gered as to fairly cover head reels and 
tail reels and both projectors in any 
theater where the print is run. No diffi- 




Fig. 2. Photomicrograph of a sprocket 
tooth of test projector. 




Fig. 3. Photomicrograph of the corner Fig. 4. Photomicrograph of a corner 
of an Eastman positive perforation. of a Dubray-Howell perforation. 



Kelley and Wolfe: Dubray-Howell Perforation 



33 



culty is foreseen in this test; in fact, 
it is expected that it, too, will support 
the results that have been obtained 
experimentally and commercially in so 
many other cases. 

The registration tests required a great 
deal of careful planning and became 
involved in factors which are not hi 
themselves a part of the test. Since the 
industry is hi the process of changing 
from nitrate base negatives to safety 
base negatives, factors which might be 
influenced by this change in base mate- 
rial could not be neglected. Similarly, 
the low shrinkage characteristics of the 
safety base film have made it necessary 
to manufacture such negatives with a 
shorter than standard perforation pitch. 
Thus the perforation pitch had also to 
be considered hi this test. 

In projection process photography, 
stationary foreground objects are com- 
monly photographed together with the 
rephotographing of a projected picture. 
This, hi the final composite result, pro- 
vides an extremely critical test of the 
steadiness of a motion picture, since the 
eye constantly has the opportunity to 
observe foreground and background 
objects where any relative motion is 
exaggerated. Since this form of photog- 
raphy presents one of the most critical 
registration problems in the industry, it 
was chosen as the test method for com- 
parison of the Dubray-Howell and Bell 
& Howell perforations. Briefly, the 
over-all process involves the first cam- 
era, a registration printer, a process pro- 
jector, a second camera, a continuous 
printer and the final theater projector. 
The outline in Fig. 5 shows the com- 
binations of all of these factors which 
were carried through in this test. The 
basic situation involved is quite simple, 
but the detail results in complications 
which require very close study for an 
understanding of the tests themselves 
and the results which have been ob- 
tained. 

The program was laid out to cover all 
practical combinations of pilot phis 



and perforations for nitrate and safety 
base, short pitch and standard pitch 
negatives, in a chain of operations which 
was as follows: A chart, as shown in 
Fig. 6, was photographed in the back- 
ground camera. The film was rewound, 
the chart shifted slightly and a second 
exposure was made. This film was de- 
veloped and printed in a step contact 
printer, resulting in a process plate as 
shown in Fig. 7. This process plate was 
then projected through a process pro- 
jector onto a translucent screen. Be- 
tween the process projector and the 
screen was a latticework covering the 
full area of the screen. This casts a net- 
work of black lines on the screen itself 
and provides a new reference point. A 
standard production-type camera photo- 
graphed this process screen and normal 
release type prints were made from that 
negative. 

Figure 8 shows a frame of the release 
print. The parallel white lines and the 
identifying slates at the top middle and 
lower right were photographed by the 
first camera and projected on the process 
screen. The network of black lines is 
the shadow of the latticework between 
the projector and the screen, the broad 
black line extending in an "L" shape at 
the lower right corner was created by a 
gobo [section of dark wallboard often 
set up to shield camera lens from light] 
placed in front of the screen. Similarly, 
the broad black line at the lower left 
corner with the narrow white stripe 
through it vertically was created by a 
gobo with a slit in it located in front of 
the screen, so that the light coming 
through the slit from the process screen 
caused the vertical white line. The 
slate hi the lower center marked "Reel 
b," was located in front of the screen. 
In projecting these prints, the salient 
points looked for were movement be- 
tween the white lines, which is a func- 
tion of the background or first camera, 
movement between the black network 
lines and the white chart lines, which is 
a function of the first camera, the 



34 



January 1951 Journal of the SMPTE Vol. 56 



STANDARD PITCH STANDARD PITCH 



FIRST CAMERA 

(SEE NOTE 316) 



STEP PRINTER 

(SEE NOTE 416) 



PROCESS PROJ. 

(SEE NOTE 516) 



SECOND CAMERA 

(SEE NOTE 7) 



RELEASE PRINT 

(SEE NOTE 8) 



PART I 

(SEE NOTE i) 

NITRATE BASE 



PART 2 

(SEE NOTE i) 

SAFETY BASE 



PART 3 

(SEE NOTE l) 

SAFETY BASE 

SHORT PITCH NEC. -STANDARD PITCH POSITIVE 




Fig. 5. Outline of perforation registration tests. 



PART 4 
(SEE NOTE 2) 
LIBRARY 



Legend: 

Numerals - Test number. 
Capital Letters - Type of perforation. 
Small Letters - Type of registration pin. 
B - Bell & Howell (negative) perforation. 
D - Dubray-Howell perforation. 
E -Eastman (positive) perforation, 
b - Bell & Howell registration pin. 
d - Dubray-Howell registration pin. 
C - Continuous printer. 

Note 1: These titles describe the film 
used in the first (background) camera and 
process (background) projector only, and 
do not refer to the second camera negative 
or the release print. 

Note 2: This is the only case where the 
background negative and positive were of 
different base material. This background 
negative was nitrate base, Bell perforated, 
exposed on a Bell & Howell pin. The 
background print was Dubray perforated, 
standard pitch, safety base, made on a step 
printer having a full fitting Bell & Howell 
pin, and projected on a background pro- 
jector having a full fitting Dubray-Howell 
pin. This would be the procedure fol- 
lowed on existing library material if the 
Dubray-Howell perforation were adopted 
as a universal standard. 

Note S: Two full-aperture first (back- 
ground) cameras were used; the one, in 
line with regular procedure, had a full 
fitting Bell & Howell registration pin; 



the other camera had a full fitting Dubray- 
Howell registration pin. 

Note 4-' Two step printers were used; 
the first had a full fitting Bell & Howell 
registration pin; the second had a full 
fitting Dubray-Howell registration pin. 

Note 5: Two background process pro- 
jector movements were used; the first had 
a full fitting Bell & Howell registration pin ; 
the second a full fitting Dubray-Howell 
registration pin. 

Note 6: The small registration pin (full 
fitting in height only) was not changed in 
any of the equipment mentioned in Notes 
3, 4 and 5. This is a satisfactory pro- 
cedure, as the Bell & Howell and the 
Dubray-Howell perforations are identical 
in height dimensions. 

Note 7: The second (rephotographing) 
camera had a Bell & Howell registration 
pin. In this particular series, the negative 
was Bell & Howell perforated, standard 
pitch, nitrate base. 

Note 8: All these release prints were 
made on a continuous printer, using an 
Eastman perforated, standard pitch, safety 
release positive. 

Note 9: This second (rephotographing) 
camera was identical to the camera de- 
scribed in Note 7, but the negative was 
Dubray perforated, short pitch, safety 
base. 

Note 10: These release prints were made 
on the same continuous printer described 
in Note 8, but using Dubray perforated, 
standard pitch, safety release positive. 



Kelley and Wolfe: Dubray-Howell Perforation 



35 



printer, or the process projector, and 
movement between the production 
camera aperture and the black gobo in 
the lower right corner, which is a func- 
tion of the production camera. Move- 
ment of the release printer is shown by 
the relationship between the sprocket 
holes and the production camera aper- 
ture line and, of course, movement in the 
final projector is shown by a movement 
of the sprocket holes themselves. 

No attempt has been made in this dis- 
cussion to enter into the fine details of 
identifying and measuring the several 
possible sources of instability, but suffi- 
cient information is provided so that by 
careful study an understanding of the 
possibilities of analyzing this chart can 
be obtained. It should perhaps be suffi- 
cient to say that by means of these 
special charts, gobos, latticeworks, spe- 
cial printer and camera apertures and 
such other devices, the source of any 
movement which takes place in this 
chart as it is projected on the screen can 



be isolated and identified. This was, of 
course, a fundamental necessity in a 
test program containing the detail in- 
volved in this one. 

Several practical problems face the 
industry if the Dubray-Howell perfora- 
tion is established as the universal 
standard; they are: projection life, the 
use of library negatives with Bell & 
Howell perforations, the use of Dubray- 
Howell perforated negatives in cameras 
with Bell & Howell pilot pins and, of 
course, the over-all optimum registra- 
tion obtainable with Dubray-Howell 
perforations throughout each step in 
the production of a motion picture. 

On the basis of these tests, the follow- 
ing predictions and conclusions are 
made: There will be no commercial 
loss in projection life of Dubray-Howell 
perforated prints as compared to East- 
man perforated prints; library nega- 
tives with Bell & Howell perforations 
can be printed to Dubray-Howell per- 
forated process plates with satisfactory 



3+0 




36 



Fig. 6. Original chart registration tests. 
January 1951 Journal of the SMPTE Vol. 56 



4(7 



RESEARCH COUNCIL 
TEST CHART 




Fig. 7. Process plate registration tests. 




Fig. 8. Release print registration tests. 
Kelley and Wolfe: Dubray-Howell Perforation 



37 



results, except where the most critical 
registration problems are involved; 
Dubray-Howell perforated negatives 
can be used in existing cameras without 
change, and release printers do not need 
to be changed for printing Dubray- 
Howell perforated negative or positive 
films, although some additional im- 
provement can be obtained if the 
sprockets in such continuous printers 
are changed to take full advantage of 
the Dubray-Howell perforation. 

The Research Council expects to 
recommend to its Board of Directors 
that the Dubray-Howell perforation 
should be presented to the American 
Standards Association for adoption as 
the universal standard perforation for 
negative and positive motion picture 
film. Furthermore, it expects to recom- 
mend to its member companies that in 
the use of Dubray-Howell perforated 
negative and print stocks for normal 
studio operations, they should change 
pilot pins in cameras used primarily for 
process background films, registration 
printers and process projector move- 
ments; a step printer with Bell & 
Howell pins should be retained for print- 
ing library negatives; the pilot pins in 
the normal production cameras will not 
need to be changed except as a matter 
of maintenance ; and release printers may 
be used without change. The Research 
Council will also recommend changes in 
pilot pins and sprockets for new cameras 



and printers, and for replacement parts. 

In support of these test results, there 
is not only the very considerable 
amount of experimental work done by 
others throughout the last twenty years, 
but currently there is considerable com- 
mercial experience. Mention has al- 
ready been made of Trucolor and Cine- 
color, both of whom are using Dubray- 
Howell perforated release prints; but in 
addition to these it should be noted that 
Eastman color positive, Du Pont color 
positive and one experimental negative, 
as well as its companion positive, are 
all using Dubray-Howell perforations. 

To summarize briefly, the tests made 
by the Research Council have confirmed 
and extended the data obtained by other 
experimenters in this field. The Re- 
search Council believes that the indus- 
try has an opportunity at this time to 
achieve the long-desired goal of a single 
standard perforation. If the Dubray- 
Howell perforation is adopted as that 
universal standard, no confusion will be 
created at any point in the industry, 
nor will it be necessary to expend any 
considerable money to make the minor 
conversions which are desirable although 
not completely necessary. 

It is strongly recommended that every 
effort be made at this time to get the 
complete support of the industry behind 
standardizing the Dubray-Howell per- 
foration for all negative and positive pur- 
poses. 



January 1951 Journal of the SMPTE Vol. 56 



Effects of Television 

on the Motion Picture Theater 

By Benjamin Schlanger and William A. Hoffberg 



The advent of television has accelerated the need for refinements and 
improvements in the art of the projected motion picture in theaters. The 
factors of cinematography, theater location, seating capacity and theater 
design have to be dealt with in accordance with circumstances which al- 
ready appear to call for a fresh approach to the problem. It is important to 
evaluate the ability to adapt existing theaters to the new requirements. 



A I/THOUGH home television seems to 
be acquiring a mass audience, 
there will always be a motion picture 
theater and theater television audience 
consisting of those patrons who wish to 
see entertainment not available hi other 
mediums, those who wish to avoid ad- 
vertising intrusions, those desiring a 
respite from the home environment, 
those satisfying their gregarious in- 
stincts and those who prefer the dra- 
matic impact of the large theater screen 
cinematography. This audience may 
be surprising in numbers because it has 
been estimated that only 10 to 20% of 
the potential audience ever attended 
even the most popular picture. 

We are now going out of a period in 
motion picture history in which great 
leeway existed in both production and 



Presented on October 20, 1950, at the 
Society's Convention at Lake Placid, N. Y., 
by Benjamin Schlanger and William A. 
Hoffberg, Theater Engineering and Archi- 
tecture Consultants, 35 W. 53d St., 
New York 19. 



exhibition. The margin for error, in- 
competence and acceptability of ques- 
tionable quality of production and ex- 
hibition is narrowing down with the 
advent of television. Now, the factor 
of quality in motion picture theater 
entertainment will determine the size 
of its audience. Of course, quality 
primarily includes story content and 
performance, but if the motion picture 
theater cannot deliver the story content 
and performance in a manner far su- 
perior to any of the other entertainment 
mediums, it will lose the mam reason 
for its existence. 

Television has accentuated the neces- 
sity for intimacy in the motion picture 
theater because each home television 
seat is a "ringside" seat. The television 
camera is located at a distance and angle 
from the scene which the director con- 
siders most favorable to the home audi- 
ence. At home, the television viewer 
has the great advantage of choosing his 
seating pattern by individual prefer- 
ence. However, the scale of the tele- 
vision screen in the home is limited. 



January 1951 Journal of the SMPTE Vol. 56 



39 



The comparatively bright illumination 
levels required in home television view- 
ing makes the viewer particularly con- 
scious of this deficiency. The inclusion 
of furniture and room details in the field 
of view does much to destroy intimacy. 
In contrast with home television, the 
motion picture theater has a fixed seat- 
ing pattern. The theater audience 
seating preferences can readily be seen 
as they choose their seats at the be- 
ginning of the show. The less desirable 
seats are then reserved for latecomers. 

Improving Theaters 

The competition of home television 
can be a healthy stimulus to induce 
theater owners to improve their physical 
plant so that the enjoyment of a motion 
picture in a theater is noticeably su- 
perior. The following items deserve 
careful consideration in this connection : 

1. All theater seat locations must be 
desirable. Unobstructed vision of the 
screen is mandatory. Ample row spac- 
ing and two arm rests for each seat will 
be necessary. 

2. The scale of the theater screen 
image should increase so that the dif- 
ference in scale as compared with the 
home television screen is accentuated 
and dramatized. 

3. Since 1938, we have advocated 
the elimination of black masking around 
the motion picture screen and we now 
have many successful installations of 
this type in theaters. The majority of 
television receiver sets have very light 
colored maskings. A luminous field 
around the screen, preferably synchro- 
nized with the screen lighting intensities, 
would reduce eyestrain and enhance 
peripheral cinematographical effects. 

4. Some of the fluidity and inventive- 
ness achieved in television production is 
worth noting. With the larger screen 
and luminous screen surround, the 
peripheral areas of the human field of 
view can be exploited for greater dra- 
matic effect. 

5. The effectiveness of distant pano- 



ramic views and medium shots on the 
television receiver is necessarily limited 
in scale. In contrast, the larger theater 
screen and the increased use and im- 
provement of wide-angle camera lenses, 
are great advantages. 

6. Development of higher intensity 
projection equipment, coated lenses, 
and the reduction of film grain as well as 
the demands of drive-in projection, 
have made larger screen projection 
feasible. 

7. Further enhancement of cinema- 
tography is produced by the increased 
subtended angle of the larger screen to 
the average viewer. 

8. Items 2 and 3 of the above recom- 
mendations can now help to bring 
three-dimensional motion pictures into 
use. With seating depth limited to 
approximately four times the picture 
width instead of the greater viewing 
depths now used, objectionable per- 
spective distortions experienced in 
stereoscopic viewing will be reduced. 
The elimination of dark picture sur- 
rounds is highly consistent with the 
realistic effect of stereoscopic viewing. 

9. Stereophonic sound in theaters 
giving positional sound effects in space 
can hardly be conceivable in home tele- 
vision sound. 

The above suggestions for improve- 
ment must, of course, be adaptable to 
existing theaters. In a survey of about 
600 U.S. theaters, which was conducted 
by this Society in 1938, an average 
screen width of 18 ft 6 in. and an average 
ratio of maximum viewing distance to 
picture width of 5.2 was found. An 
increase of average screen width to 24 
ft in. would reduce the ratio of maxi- 
mum viewing distance to picture width 
from 5.2 to 4.0 and would increase the 
screen area by about 67% . This change 
would be structurally feasible in the 
majority of existing theaters. It is 
true that in many of the existing 
theaters, the use of several of the front 
rows would be eliminated but the seat 
loss would be nominal. 



January 1951 Journal of the SMPTE Vol. 56 



With reference to the elimination of 
black screen masking, the observations 
and conclusions of L. A. Jones, S. K. 
Wolf, F. M. Falge, W. D. Riddle, B. 
O'Brien, C. M. Tuttle, R. G. Williams, 
H. L. Hogan, M. Luckiesh, and B. 
Schlanger, since 1920, have indicated 
the desirability of illumination of 
screen surroundings. The most de- 
sirable contiguous brightness has been 
found in practice to be the synchronous 
type which automatically varies with 
the brightness of the picture. Some of 
the many examples of this type are the 
Island Theater, Bermuda; Crown 
Theater, New Haven; Essoldo Theater, 
Penge, England; and the Tacna 
Theater, Lima, Peru. Further de- 
velopments and refinements for pro- 
viding a synchronous luminous screen 
surround have been incorporated into 
several theaters now under construction, 
including the Shopping Center Theater 
in Framingham, Mass., and the Bell- 
more Theater, Bellmore, L.I. 

Locating Theaters 

New motion picture theater construc- 
tion in the U.S. has not been propor- 
tional with the increase of population. 
The growth of television is probably one 
of the factors which accounts for this. 
However, new population centers and 
obsolescence of theaters, both in plant 
and location, do create a demand for 
new theaters. Several recent develop- 
ments have greatly affected the location 
and seating capacity of new theaters. 

Since 1945, new residential planning 
has tended to be in the form of large- 
scale, integrated communities very often 
decentralized. Shopping and night-life 
centers are then located either within 
the new communitiesor on the perihp- 
ery adjacent to highways. The neces- 
sities for parking areas then become a 
major consideration in theater location. 
With high land values, it is difficult for 
new theaters in existing urban night-life 
centers to provide adequate parking 
facilities. There has, therefore, been a 



tendency to locate new theaters within 
the confines of the new communities or 
in the shopping centers. 

When new theaters are located within 
the confines of new communities, they 
have the ease of accessibility of the 
neighborhood theater. The architec- 
tural planning of residential projects 
very often indicates the use of several 
smaller theaters, with capacities in the 
order of 400 to 600 seats, rather than a 
single large theater. The smaller 
theaters have fewer building code restric- 
tions and are more economical in per 
seat cost of construction. Their scale 
suggests simplicity of exterior treatment 
and amenities. They do have the virtue 
of intimacy within the interior of the 
theater and can achieve to the greatest 
degree the previous suggestions as to 
screen size and treatment. All of the 
seats can approximate the "ringside" 
seat. Availability of screen product and 
allocation of runs to groups of smaller 
theaters is an industry policy question 
of great importance. 

The location of theaters within new 
large-scale shopping centers has differ- 
ent aspects. Adequate parking facili- 
ties are available, the theater plays an 
important part in building up night 
activity and there is, generally, con- 
siderable transient automobile traffic. 
This indicates a larger capacity theater. 
To achieve intimacy in the larger theater 
is an architectural challenge. Reduc- 
tion of the interior volume of the audi- 
torium to a minimum helps to create 
acoustical intimacy. Screen size is, of 
course, increased in the larger theater 
and with it, the scale of the screen sur- 
round treatment is increased. This 
enhances the visual intimacy which is 
the prime consideration. Then, the 
shaping of walls and ceiling, the avoid- 
ance of decoration which gives scale 
"measuring rods" and the integration of 
interior lighting must attempt to ap- 
proach intimacy of space. 

New and existing theaters which offer 
to the public the seating, air condition- 



Schlanger and Hoffberg: Television and the Theater 



41 



ing, projection and sound transmission 
comforts, which are now available, and 
which add to these the increased screen 
image, the luminous screen field, the 
increased flexibility and scope of motion 
picture cinematography, the feelings of 
intimacy within the auditorium, and 
stereoscopy of sound and vision, should 
survive within the forests of home tele- 
vision antennae which have become a 
feature of the skyline. 

References 

1. B. Schlanger, "On the relation between 
the shape of the projected picture, the 
areas of vision, and the cinemato- 
graphic technic," Jour. SMPE, vol. 24, 
pp. 402-409, May 1935. 

2. B. Schlanger, "Motion picture audi- 
torium lighting," Jour. SMPE, vol. 34, 
pp. 259-264, Mar. 1940. 

3. B. Schlanger, "A method of enlarging 
the visual field of the motion picture," 
Jour. SMPE, vol. 30, pp. 503-509, 
May 1938. 

4. M. Luckiesh and F. K. Moss, "The 
motion picture screen as a lighting 
problem," Jour. SMPE, vol. 26, pp. 
578-591, May 1936. 

5. Ben Schlanger, "Auditorium light con- 
trol through surface treatment," Better 
Theaters, Mot. Pic. Herald, Apr. 2, 1938. 

6. Ben Schlanger, "Increasing the effec- 
tiveness of motion picture presentation," 
Jour. SMPE, vol. 50, pp. 367-373, 
Apr. 1948. 

Discussion 

PIERRE MERTZ: Some years ago, there 
was a development in films which seemed 
to cover something of what Mr. Schlanger 
had in mind with regard to the wide screen 
the Grandeur film. That occurred be- 
fore I came into this field. Can you tell 
us, what was the improvement in realism 
with the Grandeur film as compared with 
the conventional film? 

MR. SCHLANGER: There are many 
factors involved. First, there was a larger 
physical width of film, and I believe since 
then the film grain problem has been more 
or less licked and that a sufficiently large 
picture can be projected from 35-mm 
width. The present standard gives a 
wide enough picture in theaters, and the 



real problem, which was not licked at the 
time that Grandeur and other wide, en- 
larged screens were presented, was the 
cinematographic problem. It is quite 
natural. It was a new tool and it never 
had its chance for the experience or prac- 
tice that is needed with a new tool. In 
other words, the cinematographers never 
became familiar with the new tool or its 
potentials at that time. Today we are in a 
spot where we know we need some new 
method or device, and, should we find it. 
the cinematographers will learn to use it. 
As to the realism that can be achieved, 
there is another problem in addition to 
that of the size of film and the art of cine- 
matography that is the taking-lens in the 
camera. I remember getting in touch with 
some of the authorities and manufacturers 
of lenses to try to find out why there were 
not wider-angle lenses available or used in 
taking motion pictures, and the significant 
answer was that there was never any great 
demand for them. But it was possible to 
develop them. I do hope that they will 
develop wider-angle lenses, because that is 
another tool in the flexibility of cinema- 
tography that is necessary. 

FREDERICK J. KOLB, JR.: Most of the 
desirable features of theater design that 
you have discussed seem directly contrary 
to the requirements of a drive-in theater. 
Is it possible to reconcile the two? 

MR. SCHLANGER: Would you be specific 
as to their being contrary? 

DR. KOLB: I am thinking of the drive- 
in theater as having a very limited angle of 
view more like the home television view- 
ing conditions. Therefore the advantage 
to be gained by including a larger story 
element on the screen and by restricting the 
audience to the most favorable locations 
seems very difficult at least, to me to 
realize in drive-in design. 

MR. SCHLANGER: In drive-in theaters, 
the remote car positions are at least 10 W 
[W = screen width]. They are placed so 
because of the physical problem of getting 
enough attendance with one screen and I 
have noticed that there have been some 
developments recently for double screens 
and even four screens. I guess that is one 
of the problems to be overcome. From a 
10 W location in a drive-in theater, the 
picture looks like a postage stamp. It is 
not that it is poorly done. It is an incon- 



42 



January 1951 Journal of the SMPTE Vol.56 



spicuous speck in the field of view. How- 
ever, the drive-in theater is a unique ex- 
perience to be able to ride out in your car 
and go and view a picture is still "some- 
thing different." The audience will toler- 
ate a lot when a thing is unique enough. 
For example, even home television, good as 
it is today, falls far short of the quality of 
a motion picture in a theater. But it is 
tolerated; it is considered all right because 
it is unique. You can sit in your slippers, 
smoke a cigar and watch television without 
leaving your house. Getting back to your 
question can you produce a picture which 
is just as useful in a drive-in theater as in 
any other theater? There is an inconsist- 
ency in this respect and it can be related 
also to television viewing. Due to the 
deficiencies in television viewing there is a 
tendency, and justifiably so, to use close- 
ups, because middle and distance shots 
appear indistinct. For the same reason, 
middle and distance shots in drive-in 
theater production should also be avoided. 
There again, a predominance of close-up 
shots is a desirable thing, if drive-in thea- 
ters are going to be designed with 10 W 
viewing. So, you are correct. A picture 
which would be photographed carefully for 
a drive-in would not be good for regular 
motion picture theaters, but there is always 
a happy medium. You must be sure that 
the close-ups are not too close up, and 
that the distant shots are not too distant. 
You have to compromise, and I believe 
that this could be done easily enough so 
that there would be neither too many 
close-ups for viewing in the regular theater, 
nor too few, for the drive-in theater. 

WALTER E. DUNN: You have made re- 
peated references to the elimination of 
black screen masking. Do you have any 
recommendation for either a substitute or 
a system of elimination of the mask in an 
existing thaeter? 

MR. SCHLANGER: There are several 
methods of eliminating black masking. 
First of all we have to realize that black 
maskings were originally created for pur- 
poses which no longer exist. One was that 
screen illumination in the early days was 
comparatively low and the black masking 
went a long way toward making the illu- 
mination appear brighter. I think that 



television viewing is proving that black 
masking is no longer necessary. With the 
exception of the Du Mont sets, practically 
all the sets have a white or almost-white 
color masking. The other reason for black 
masking was to do something about the 
aberrated or fuzzy edge of the picture as 
it is when projected without a black 
masking. That is a practical problem. 
This aberrated, fuzzy edge can be elimi- 
nated in several ways. We have been de- 
veloping a substitute masking, a luminous 
masking, which I think will be available 
very soon. We have also had other solu- 
tions in which we would cut the picture, 
that is, project the picture very carefully 
into a proscenium which was exactly 
the size of the picture and let it go at that, 
or by having a slight flare come right out 
from the picture. The fuzzy edge would 
fall on the angular surface, which would 
not be visible to the audience, and the pic- 
ture would appear to have a clean-cut edge. 
Some of the newer maskings that have 
been developed will do an even better job. 

LEONARD SATZ: I'd like to add to that 
that there are certain things which, in my 
opinion, can be done right now, short of 
making major changes. I would say, prin- 
cipally, modernization of lighting would be 
the first step in the theater auditorium 
the elimination of distracting side-wall 
brackets, which are so common in many of 
our theaters, and replacement with an 
operating light which is directed downward 
and perhaps intentionally directed to the 
proscenium area. The first step would be, 
naturally, the enlargement of the screen, 
and I believe it is a fact that visual acuity 
is not lost by the reduction in screen 
brightness as long as the image is in- 
creased in size. You mentioned limitation 
of screen brightness as being one of the 
problems of the exhibitor today. I think 
that if he does lose 10% in incident illu- 
mination by enlarging his picture with 
existing projection equipment, the loss will 
be compensated by the fact that visual 
acuity is maintained with the larger picture. 

MR. SCHLANGER: It may not be exactly 
compensated, but certainly acuity in- 
creases with the size of the image, despite 
loss in light. I don't have exact fig- 
ures on that, but I believe you can verify it. 



Schlanger and If off berg: Television and the Theater 



Some Comparative Factors of Picture 
Resolution in Television 
and Film Industries 

By H. J. Schlafly 



This paper reviews and compares the quantitative meaning of the term 
resolution as commonly used by the television industry and the film indus- 
try. The danger of using values of limiting resolution as the sole measure 
of picture quality is discussed. Conversion equations are developed and 
tables listing numerically equivalent values of resolution are provided. 



THE MERGER of electronics and 
photography into the corporate 
function of television recording has re- 
sulted in a unique situation. It is a 
situation which is logical and natural, 
but which, nevertheless, has caused 
misunderstandings, delays and even 
exasperation. The problem is simply 
one wherein two sciences that have 
hitherto been comparatively independ- 
ent of each other suddenly find that 
they define and describe certain phe- 
nomena in terms which are not identi- 
cal, but which are similar enough to be 
thoroughly confusing. 

The ultimate objective of both tele- 
vision and photography is the faithful 
reproduction of an original scene. 
But, while the beginning and end 
products are the same, the medium and 
methods are widely different. Thus, 
it is little wonder that there are few, if 

A contribution by H. J. Schlafly, Twen- 
tieth Century-Fox Film Corp., 444 W. 
56th St., New York 19. Reprinted from 
Proceedings of the I.R.E. for January 1951. 



any, existing experts who are so thor- 
oughly familiar with the terminology 
and techniques of both sciences that 
they can point out in advance the areas 
of confusion or misunderstanding. This 
paper will attempt to deal with only 
one "area of confusion," the meaning of 
picture resolution as defined by termi- 
nology in current use. 

General 

The resolving power of a medium or a 
device is a measure of the ability of 
that device to convert, transmit or re- 
produce details of the original scene. 
Detail, of course, is a "separately con- 
sidered particular," 1 which contributes 
to or is part of the whole. A device 
which is capable of handling more or 
finer detail is said to have the greater 
resolving power and the resulting 
picture has more resolution. Lack of 
picture resolution not only results in 
the subordination or complete loss of 
parts of the original, but also in a loss of 
"edge sharpness" which gives the pic- 



January 1951 Journal of the SMPTE Vol. 56 



ture a "soft" or, more correctly, a 
diffuse quality. The accepted method 
of determining resolution is to provide a 
scale or chart having calibrated points 
or steps of increasing fineness of detail 
and to determine the point at which the 
device under test breaks down in the 
performance of its function. The hu- 
man eye itself has a certain resolving 
capability which is influenced by the 
portion of the retina being used, the 
spectral content of the light and the 
absolute value of the light energy, as 
well as by the optical characteristics of 
the lens. Each technical device which 
precedes the seeing process of the eye 
has its own resolution characteristic 
and contributes its part to the degra- 
dation of the original scene. 

In general, the deterioration con- 
tributed by any physical device is evi- 
denced by a gradual reduction in con- 
trast ratio with increasing detail until 
a point is reached where there is no dis- 
tinction between two adjacent points 
which did have some quality of distinc- 
tion in the original. Whether this con- 
trast ratio is measured in light energy, 
grains of silver deposit per area, poten- 
tial difference or whatever, is imma- 
terial. A notable exception to this 
gradual deterioration of resolution is the 
"sharp cutoff" voltage amplifier which 
might maintain constant amplification 
with increasing frequency (detail) until 
a certain critical or cutoff point is 
reached, and thereafter drop sharply 
toward zero output. 

Comparative physical sizes play a 
large part in determining the point 
where "signal attenuation" begins to 
occur. Thus, so-called "aperture size," 
or the area within which there can be no 
differentiation, such as a single nerve 
ending in the eye, the focused scanning 
spot in a cathode-ray device or the grain 
size in a photographic emulsion, is a 
major contributor to the limitation of 
resolution. But there are many other 
contributing causes which do not neces- 
sarily deal with physical size, such as 



electrical time constants; aberrations 
in optical devices; phase shift in ampli- 
fiers; spectral sensitivity of emulsions, 
photocathodes and lenses; and, un- 
fortunately, others. 

Today both the photographic and 
television industries speak of the abso- 
lute limit of resolution as a measure of 
picture quality. Actually the evalu- 
ation of quality is so complex that 
measurement of one of the contributing 
factors is not adequate to describe the 
end result. Much work has been done 
and is being done to determine all of the 
factors involved. 2 - 3 In particular, ana- 
lytical attention is being given to detail 
contrast ratio, random noise, bright- 
ness, and tone reproduction as well as 
to limiting resolution. The paragraphs 
which follow deal only with definitions 
and conversion factors for the resolution 
terminology in current use and should 
definitely not be considered as the sole 
measure of picture quality. 

Terms 

One is likely to assume that the use of 
the common term "lines" permits a 
basis for comparison between photog- 
raphy and television picture resolution. 
Such is not the case. Each industry 
has independently arrived at a defini- 
tion in language best suited to its own 
measurement technique and, as a result, 
numerical values which are not appar- 
ently related might refer to the same 
degree of "absolute" resolution in a 
television picture and in a photograph. 

The film industry defines resolution in 
terms of lines/mm of film surface. 
Typical test charts are provided by the 
National Bureau of Standards (shown 
in Fig. 1) and by the American Stand- 
ards Association. Such charts usually 
consist of a series of blocks or squares of 
parallel black lines separated by clear 
spaces of the same width. Each block 
represents a given number of black 
lines/mm of film surface when the chart 
is photographically reproduced on the 
film emulsion. For determining resolu- 



H. J. Schlafly: Television and Film Resolution 



45 




= 11 



NATIONAL BUREAU iOF STANDARDS 
TEST CHART | 25 X 

Figure 1. 

tion values given in film specification 
sheets, the contrast ratio between the 
black lines and the clear spaces on the 
original chart is held at 30:1. Inci- 
dentally, this is about the highest value 
which can be obtained on a printed 
chart. Transmission-type charts, used 
in some resolution measurements, can 
provide contrast ratios of 100:1 or 
1000:1. 

The resolving power of a given film 
emulsion is determined by photograph- 
ing a test chart using the optimum ex- 
posure, processing the film by recom- 
mended methods and examining the 
image under a microscope. The maxi- 
mum number of black lines/mm just 
resolved, not lost as an indistinguishable 
gray mass, is the value used to indicate 
the resolving power of that particular 
film. In practice the resolving power 
values of commercial films vary from 
about 55 lines/mm for negative film to 
as high as 150 lines/mm for fine-grain 
sound recording films. 

Of course, the figures given in the 
above paragraph do not necessarily 



represent the end product of film resolu- 
tion as seen on the screen of a motion 
picture theater. In February, 1946, a 
portion of the Television Committee of 
the Society of Motion Picture Engineers 
made observations of screen resolution 
of a special test film projected in a 
group of leading New York theaters. 
These data were not published because 
the tests were not sufficiently extensive 
to permit definite conclusions. In the 
words of the Committee report: "The 
influence of many individual factors has 
not been determined, but it is believed 
that the results. . .are broadly repre- 
sentative of present motion picture prac- 
tice " The conclusion reached in 

the same report stated, "In general, it 
can be concluded from theater projec- 
tion of the two test films specially pre- 
pared for the use of this Committee that 
projection in first-run theaters shows 
resolution of 28 lines/mm on 35-mm 
film where the test object includes pic- 
torial subject matter and 40 lines/mm 
where the test card alone was photo- 
graphed." 

In the television industry picture reso- 
lution is usually measured with the aid 
of a test pattern such as the RMA 
Resolution Chart 1946. This chart 
follows the practice of using horizontal 
and vertical wedges rather than a series 
of parallel lines. The pattern is com- 
posed of a given number of alternate 
black and white lines of equal width 
which continuously converge from the 
wide to the narrow end of the wedge. 
Thus, the chart is provided with a con- 
tinuously variable resolution pattern, 
numerically calibrated by indexing 
various points along the wedge. Each 
black and each white line is counted as 
an individual line, whereas in the film 
industry each black line only is counted 
as an individual line. 

The resolution of the television pic- 
ture is indicated by a value which repre- 
sents the limiting number, of black and 
white lines identifiable as such, not lost 
in an indistinguishable gray, in a verti- 



January 1951 Journal of the SMPTE Vol. 36 



Condition of Plot: 

Integrated Value Within 20% of Original 




0.7 

Width of Scanning Line 
Scanmng Line Factor = W idth of Vertical Detail 



Fig. 2. Integrating effect of scanning line on vertical television resolution. 



cal or a horizontal dimension equivalent 
to the picture height. For the purpose 
of assigning this value, it is assumed 
that the resolution of any and every 
point in the picture is equal to that 
observed at the wedge. Such an as- 
sumption is, of course, not true, but it is 
a convention which provides a numerical 
value of resolution accepted throughout 
the industry. Degradation at the cor- 
ners of the picture sometimes is identi- 
fied by the term "corner resolution" and 
is evaluated by the same process inter- 
preting the resolution of a wedge located 
in the corner in terms of the full dimen- 
sion of picture height. 

It is a common error to confuse the 
number of horizontal scanning lines as 
set by the television standards with the 
figure for picture resolution. The tele- 
vision standards in this country specify 
525 horizontal scanning lines per picture 
frame. Only 92% to 95% of these are 
active scanning lines, the remainder be- 
ing blanked out during the vertical 
sweep retrace. But even the remaining 
480 some odd lines do not specify the 
limit of vertical resolution. There is an 
additional loss in vertical resolution in- 
herent in the television dissecting proc- 



ess which provides a second factor even 
when there is perfect interlace of the 
alternate scanning fields. This effect is 
illustrated by integration of those por- 
tions of black and white resolution lines 
within the width of the scanning line 
and point by point comparison of the re- 
sulting halftone with the original. 4 
Figure 2 plots such information for a 
range of scanning line factors, showing 
the percentage of possible scanning- 
line-resolution-line phasing for which 
the integrated halftone will be within 
20% of the original. Choice of a scan- 
ning line factor may be a matter of 
individual preference but a value of 0.75 
is commonly accepted and is the value 
used in the equation derivations in- 
cluded in this paper. 

Using these factors, present-day 
standards, therefore, impose a limitation 
on vertical resolution of the television 
picture of approximately 360 lines. 

Television picture resolution, by vir- 
tue of common usage among electronic 
personnel, has also come to be identified 
in terms of bandpass, or maximum pass 
frequency of the video circuits. Such 
usage has meaning only when applied to 
horizontal resolution and then only if a 



H. J. Schlafly: Television and Film Resolution 



47 



definite horizontal scanning period is 
specified. One cycle of a particular 
video frequency during active horizontal 
scanning represents one dark and one 
light picture element on a particular 
scanning line. The higher the video 
frequency, the greater the number of 
picture elements that can be theoreti- 
cally squeezed into one line. Ideally 
the one cycle which supplies the light 
and the dark picture element should 
contain sufficient harmonics to resemble 
a square corner pulse; actually, a 
sinusoidal waveform is considered suffi- 
cient for the limiting condition, sacri- 
ficing "edge sharpness" between picture 
elements. 

It will be realized that a longer scan- 
ning period would permit more cycles of 
video signal to be included in one scan- 
ning line and thus the value of hori- 
zontal resolution would be increased. 
The scanning period is set by the hori- 



zontal scanning frequency, or, by the 
combination of picture frames per 
second and scanning lines per frame. 
Figure 3 indicates the relationship be- 
tween video bandpass and horizontal 
resolution for several values of scanning 
lines per frame. 

It is interesting to note that the video 
bandpass of 4.5 megacycles, the nominal 
television broadcast standard, results in 
a horizontal resolution of approximately 
360 lines. Thus, it is seen that the 
present standards provide about the 
same picture resolution in the vertical 
and horizontal coordinates. 

Conversion Factors 

A. Conversion of Film Resolution in 
Lines per Millimeter to Television Resolu- 
tion in Lines. 

Rt = %H/Rf 

where Rt = television resolution in lines 
per picture height 



700 



Scanning, Lines per Frame./ 



Television Broadcast 
Standard U.S.A. \ 



X Condition of Equal Vertical 
and Horizontal Resolution 



Frame Rate 30/Sec. 
Vertical Blanking 0.065 
Horizontal Blanking 0.17 
Line Scanning Factor 0.75 




48 



6 8 10 12 14 

Maximum Video Frequency in Megacycles 

Fig. 3. Television resolution. 
January 1951 Journal of the SMPTE Vol. 56 



Rf = film resolution in lines /mm 
Hf = height of standard motion 

picture projector aperture in 

millimeters. 



For 35-mm film H/ 
Rt 



15.25 mm 
30.5 Rf 



For 16-mm film H/ = 7.21 mm 
Rt = 14.42 R f 

B. Conversion of Television Scanning 
Lines per Frame to Lines of Vertical 
Resolution (Television). 

R tv = b(aL) 

where R tv = vertical resolution (televi- 

sion) in lines 

a = vertical blanking factor 
b = line scanning factor 
L = total number of scanning 

lines per television frame. 

Substituting present standards: 

a = 0.92 min., 0.95 max., 0.935 

average 

6 = 0.75 (representative) 
L = 525 lines 

R tv = 0.701 L = 0.701 X 525 = 
368 lines. 

C. Conversion of Maximum Video 
Pass Frequency to Lines of Horizontal 
Resolution (Television). 



Rtk = 2/ max n/A 

= horizontal resolution (televi- 

sion) in lines 
= maximum video pass fre- 

quency in megacycles 
active time (unblanked) of 

horizontal sweep in micro- 

seconds 



where Rth 

/max 



where C = horizontal blank- 
ing factor 

F r = frames per sec- 
ond (television) 

A = television aspect 
ratio. 

Substituting present standards: 

/max = 4.5 megacycles 

C = 0.82 min., 0.84 max., 0.83 

average 
F r = 30 frames/sec 



then Rth = 2 X % X 52.7 X 4.5 
= 79 X 4.5 = 356 lines 

or general formula 

Rth = 79 /max- 

D. General Conversion Formulas for 
Equal Resolving Power Between Film and 
Television. 

1. Television scanning lines per frame 
in terms of film resolution (required for 
equal vertical resolution) : 

L = (2/ab)H f X Rf 

2 
= 0.935 X 0.75^' X Rf 

L = 43.5 Rf for 35-mm film 
L = 20.6 Rf for 16-mm film. 

2. Maximun video frequency in terms 
of film resolution (required for equal 
horizontal resolution) in 525-line, 30- 
frame television system: 

/max = (jj X L X F r X Hf X IQ~*\ Rf 

3^3 X 525 X 3 X 10 ~ 6 X 



fmax = 0.386 Rf megacycles for 35-mm 

film 

= 0.182 Rf megacycles for 16-mm 
film. 

3. Maximum video frequency in 
terms of film resolution (required for 
equal horizontal resolution in a 30- 
frame television picture), if the number 
of scanning lines in that picture has 
been chosen to give equal vertical reso- 
lution: 



A = 4/3. 



/max = 0.032 Rf* megacycles for 35-mm 

film 

= 0.00715 Rf 2 megacycles for 16-mm 
film. 

The above equations have been 
applied to several values of film resolu- 
tion for both 35-mm and 16-mm sound 
film and the results have been tabulated 
in Tables I and II. These tables list 



H. J. Schlafly: Television and Film Resolution 



49 



Table I. 35-Mm Sound Film 



Numerically 

Equivalent Values 

of Resolution 



Minimum Television 

Standards Required for 

This Resolution 



Film Hori- Vertically & 

(lines Tele- zontally* Horizontally 

per vision (video (lines/ (video 

mm) (lines) freq.) frame) freq.) 



90 


2740 


35 me 


3900 


260 me 


40 


1220 


15 


1700 


51 


28 


850 


11 


1200 


25 


17 


520 


6.5 


750 


9.3 


11 


335 


4.2 


475 


3.9 



Table II. 16-Mm Sound Film 



90 


1300 


16 me 


1850 


58 me 


40 


580 


7.3 


820 


11 


28 


400 


5.1 


580 


5.6 


17 


250 


3.1 


350 


2.1 


11 


160 


2.0 


230 


0.9 



* Provided the standard of 525 scanning lines per 
frame is retained. 

Note: When transcribing film to television or 
television to film, degradation factors of each 
system are cumulative. To minimize over-all 
degradation the resolution capabilities of one 
system should substantially exceed that of the 
other. The magnitude of this "safety factor" 
is governed by operational techniques. 



numerical equivalent values of resolu- 
tion and the corresponding television 
standards which would be necessary to 
realize such a value first, of horizontal 
resolution (with 525 scanning lines per 
television frame, 30 frames/sec) and 
second, of both vertical and horizontal 
resolution (with 30 frames/sec). 

These tables could be interpreted to 
say that, provided all other factors affect- 
ing picture quality are equal, a television 
picture having a limiting resolution of 
360 lines (the approximate capabilities 
of the existing television broadcast 
standard in the United States) is equiva- 
lent to a 35-mm sound motion picture 
film having a limiting resolution of 
about 12 lines/mm; or, to a 16-mm 
sound motion picture film having a 
resolution of about 25 lines/mm. In 
actual practice film resolution having a 
limiting value of 30 to 40 lines/mm is 
not difficult to achieve but, on the other 
hand, the reproduced film picture is not 
able to maintain the contrast ratio that 
can be realized in a reproduced tele- 



vision picture as detail approaches the 
television cutoff value. Some workers 
in the field believe that the "other fac- 
tors affecting picture quality" men- 
tioned above may eventually be so im- 
proved in the television system that ex- 
isting standards will permit a television 
picture quality closely approximating 
that of the present-day 35-mm motion 
picture film in spite of wide differences 
in the limiting value of picture resolu- 
tion. 

It must be emphasized again that the 
tables provide numerically equivalent 
values of resolution. They do not in 
themselves permit a comparison of pic- 
ture quality. They in no way indicate 
the film resolution that is required when 
a film is to be reproduced over a tele- 
vision system or when a television pic- 
ture is to be reproduced on film. It is 
obvious that when a film is reproduced 
by a television system, or vice versa, 
the end result will contain the defects of 
both. For best results, therefore, both 
systems should be operated as close as 
possible to their limit of perfection, or, 
in some cases, be controlled to com- 
pensate for defects or limitations of the 
other. 5 

Summation 

Picture quality and picture resolution 
are not necessarily synonymous. A 
figure indicating picture resolution is 
generally a numerical measure of the 
limit of detail distinction. Picture 
quality is a function not only of the 
limit of detail distinction, but also of the 
attenuation characteristic which accom- 
panies the reproduction of increasing 
detail, and numerous other factors of 
reproduction. 

The film industry speaks of resolution 
as a figure indicating the maximum 
number of black lines, separated by 
white spaces of equal width, which can 
be identified in a dimension equal to one 
millimeter of film surface. 

The television industry speaks of 
resolution as a figure indicating the 



50 



January 1951 Journal of the SMPTE Vol. 56 



maximum number of alternate black Rev., vol. 9, nos. 1-4, pp. 7-34, 246-282, 

and white lines of equal width, which 491-527 and 653-686; Mar., June, 

can be identified in a dimension equal Sep*- and Dec - 1948 - 

to the picture height. 4 - A detailed analysis of the effect of 

finite scanning apertures can be found 

References in: P. Mertz and F. Gray, "Theory 

1. Funk & Wagnalls New College Standard of scanning and its relation to the char- 
Dictionary. acteristics of the transmitted signal in 

2. M. W. Baldwin, Jr., "The subjective telephotography and television," Bell 
sharpness of simulated television im- Sys. Tech. Jour., July 1934. 

ages," Bell Sys. Tech. Jour., vol. 19, 5. C. L. Townsend, "Specifications for 

pp. 563-587, Oct. 1940. motion picture films intended for tele- 

3. O. H. Schade, "Electro-optical charac- vision transmission," Jour. SMPTE, 
teristics of television systems," RCA vol. 55, pp. 147-157, Aug. 1950. 



H. J. Schlafly: Television and Film Resolution 51 



Image Tubes and Techniques 

in Television Film Camera Chains 

By R. L. Garman and R. W. Lee 



In this country the iconoscope is used almost universally for motion pic- 
ture film camera chains. In Europe the flying-spot scanner has recently 
come into extensive use. Other pickup devices, storage and nonstorage, 
such as the image orthicon, image iconoscope and image dissector tube, 
have been used experimentally or in a limited commercial way. The 
characteristics of each of these tubes and their associated equipment are 
discussed, and certain advantages are evaluated with respect to such factors 
as signal-to-noise ratio, spurious signals, spectral response and transfer 
characteristic. 



SINCE THE TIME of the early mechan- 
ical schemes of light-spot scanning, 
many different techniques have been 
employed for producing television sig- 
nals from pictures on film. A relatively 
short while ago, charge-storage tubes 
were acclaimed as a great step forward 
and away from the rotating-disc or 
rotating-drum mechanical scanners. 
Electronic techniques which are now be- 
ing advanced as a desirable substitute 
for charge-storage tubes are exactly 
analogous to the early mechanical 
schemes. Historically, one develop- 
ment cycle seems to be complete. It is 
not only possible, but very probable, 
that further development will produce 
significant new advances in the art. A 
review of film projection methods and 
equipment now in use here or abroad 



Presented on April 25, 1950, at the Soci- 
ety's Convention at Chicago, by R. L. 
Garman and R. W. Lee, General Precision 
Laboratory, Inc., Pleasantville, N.Y. 



seems very much worth while at this 
time. 

In the discussion which follows, only 
those projector mechanisms which use a 
single film and a single gate are consid- 
ered. More complex schemes which 
have been proposed are omitted, not 
through lack of merit, but because space 
does not permit their inclusion. Also, 
the survey of photosensitive image 
tubes is restricted to those commercial 
types which are currently available. 

Projector Mechanisms and Timing 
Diagrams 

The timing diagram of Fig. 1 indicates 
the nature of the basic requirements on 
the projector mechanism. The tele- 
vision vertical sweep and retrace are 
displayed for reference. 

The television field frequency and the 
projector frame rate are those common 
to American practice, which is charac- 
terized by a conventional 24-frame/sec 
projector rate and a standard 60 cycle/ 



52 



January 1951 Journal of the SMPTE Vol.56 



sec vertical sweep frequency. It may 
be noted that British and continental 
European television practice is based on 
a 25-frame/sec sweep standard. Con- 
ventional film, recorded at 24 frames/ 
sec, provides very satisfactory results 
when played back frame for frame at 
this sweep frequency. The simplicity 
of frame-for-frame playback is not 
possible in this country, where the clos- 
est sweep rate that can be used for play- 
back is 30 frames/sec. Fortunately, 
the television and motion picture frame 
frequencies are commensurable, with the 
motion picture frame time of 1/24 sec 
corresponding exactly to that of two 
and one-half television fields. 

The usual method employed to make 
up for the difference in frame rates is 
that of scanning one film frame twice, 
the next three times, the next one 
twice, etc. Pulldown timing diagrams 
which accomplish this "2-3-2" method 
of scanning are illustrated in the last 
three lines of the timing diagram. The 
shaded areas correspond to intervals of 
time during which the film is in motion. 



Exposure of the pickup device to 
light from the film entails three tasks. 
The first is the transition from one mo- 
tion picture frame to the next. The 
second is illumination of the field, which 
must not take place while the film is in 
motion. The third is a raster scanning 
process, which may be accomplished by 
any one of several different methods. 
Some of these methods require that 
scanning be completed during the 
illumination of the field; others permit 
illumination of the field during a part of 
the scanning interval; still others do not 
permit illumination during any part of 
the scanning interval. Figures 2 to 4 
provide a more detailed breakdown of 
the basic timing diagram with regard to 
these differences. 

Figure 2 shows timing diagrams for 
film pickup systems in which the field is 
illuminated during the television sweep 
retrace time. Again, the vertical sweep 
and the vertical retrace period are shown 
for reference purposes. Illumination 
during the television sweep retrace time 
can be used with any storage-type 



"REALLY FAST" 
PULLDOWN {< 8) 




.OR. 

FRAME DISSOLVE IN 
NON-INTERMITTENT 
PROJECTOR 



144* SHUTTER ROTATION 



Fig. 1. Basic timing diagram for 24-cycle 
television projector mechanisms. 



Carman and Lee: Film Camera Chains 



53 




OR 

2-3-2 12 CPS I 
PULLDOWN 
(<I25M 



CAN BE USED WITH ANY 
STORAGE-TYPE IMAGE TUBE 



Fig. 2. Film pickup timing diagram; illumination 
during television sweep retrace. 



VERTICAL SWEEP 



VERTICAL RETRACE 



"FAST" 
2-3-2 12 CPS 
PULLDOWN 



II 

eo CPS ILLUMINATION'! 



PATTERN FOR ABOVE 



"FAST" (<55) 
24 CPS PULLDOWN 



120 CPS ILLUMINATION 
PATTERN 



rtn 




NOTE: EITHER 1440 RPM OR 1800 RPM 
SHUTTER MAY BE USED. 



CAN BE USED ONLY WITH 
INTEGRATING STORAGE TUBES 



Fig. 3. Film pickup timing diagram; illumination 
and pulldown during television sweep. 



54 



January 1951 Journal of the SMPTE Vol. 56 



image tube such as the iconoscope, the 
orthicon, the image iconoscope and the 
image orthicon. The relationships illus- 
trated are typical of the mechanisms 
now in use in iconoscope film camera 
chains in this country. Two types of 
pulldown timing are in use, the standard 
24-cycle/sec "fast" pulldown, and the 
"2-3-2" mechanism with a basic repeti- 
tive pattern at 12 cycles/sec, in which 
nearly the full television field period is 
available for film motion. 

Figure 3 shows timing diagrams for 
film pickup systems in which both illumi- 
nation and pulldown of film may occur 
during the television sweep time. This 
method can be used only with a storage 
tube which integrates linearly the light 
failing upon the photocathode. The 
only example of such a tube at present is 
the image orthicon, which may be used 
quite successfully for this application. 
Of course, the film may not be illumi- 
nated while in motion, or "travel 
ghost" will result, as in any intermittent 
projector without a shutter. The two 
bottom lines illustrate the timing for one 
very simple scheme, which may be used 
with projectors which have a satisfac- 



torily fast pulldown (60 or less). The 
120-cycle/sec illumination pattern for 
this scheme is generated very simply 
by a regular 24-cycle/sec shutter with 
five equally spaced slots. Unfortu- 
nately, there are extraneous photoelec- 
tric effects hi the image orthicon which 
limit the minimum exposure time for 
this kind of operation. These effects 
are often visible as a streak across the 
picture, called an "application bar." 
The visibility of this bar is more or less 
proportional to the peak illumination. 
It is therefore advisable to increase the 
duty cycle of the projector. There is no 
strict limit, but generally the perform- 
ance is acceptable if the shutter open 
angle is greater than 30. A satisfac- 
tory solution is the use of a rather fast 
"2-3-2" pulldown mechanism with a 60- 
cycle/sec illumination pattern obtained 
from a 12-, 30- or 60-cycle/sec shutter. 
Inspection of the diagram will show that 
if the pulldown time is approximately 
50 in the "2-3-2" mechanism, exposures 
of 70 to 80 of shutter rotation can be 
obtained. 

Figure 4 shows timing diagrams for 
film camera chains in which pulldown 



VERTICAL 
SWEEP 


II ' 

H/l/i/ 


iiii 

\AAAA 


VERTICAL 
RETRACE 


n 
H 
II 

I 


6 ^SEC.* 

1 11 


\ 
\ 

\ n n 


I 


ILLUMINATION 


ii 

In 


n 


* *> 


n 


nrin 
i 1 


rj 
i i 


PULLDOWN 
OR 
FRAME DISSOLVE 


n 








CAN BE USED WITH ANY j 
PICKUP DEVICE^ STORAGE 
OR N6N- STORAGE. 

1 1 1 
1 1 1 



Fig. 4. Film pickup timing diagram; pulldown 
during television sweep retrace. 



Garman and Lee: Film Camera Chains 



55 



occurs during the television sweep re- 
trace time. This mode of operation can 
be used with any photosensitive pickup 
device, storage or nonstorage. The 
illumination must properly be inter- 
rupted every 1/60 sec in order to provide 
equal exposures for each television field 
scan. If a nonintermittent projector 
with lap dissolve from one frame to the 
next is used, continuous illumination is 
possible. It may well be that, even in 
the case of the intermittent projector, 
film travel might be so fast that " travel 
ghost" would not result from illumina- 
tion during the retrace time. This 
possibility may have more than merely 
academic interest. Ten years ago most 
engineers were convinced that an inter- 
mittent projector with pulldown during 
the retrace time was not only impossible, 
but fantastic and ridiculous. This is 
not the case today. Mechanical inter- 
mittent mechanisms which are simple 
extrapolations of conventional design 
are now available, and will pull down 
film in approximately 15 of shutter 
rotation. Audible noise problems are 
often acute with these mechanisms. 
Completely new approaches to the 
problem now give promise of providing 
pulldown in less than the minimum 
vertical retrace standard set by the 
Federal Communications Commission! 

Tube Characteristics 

The film projector and the image- 
sensitive tube are the two elements 
which distinguish film operations from 
live studio techniques. In general, the 
tube types used for film pickup are the 
same as those developed for live pickup. 
They are the photomultiplier (used in 
conjunction with a flying-spot scanner), 
the image dissector (preferably with an 
electron multiplier), and the several 
storage tubes including the iconoscopes, 
image iconoscopes, orthicone and image 
orthicons. The parameters which are 
important in the selection of a tube are 
Hignal-to-noise ratio, transfer character- 
istic, freedom from spurious signals, 



spectral response and sensitivity. Res- 
olution capability is equally important 
but will be disregarded in this discussion 
because of the lack of good data on 
which to base conclusions. 

It is extremely important to recognize 
that the signal-to-noise ratio and the 
shape of the transfer characteristic can- 
not be considered independently in 
arriving at a real evaluation of obtain- 
able picture quality. The signal-to- 
noise ratio, measured as the ratio of peak 
signal to rms noise, is very much affected 
by the nature of the transfer character- 
istic of the over-all system, as well as by 
the noise distribution over the range of 
light flux utilized. Schade 1 has pro- 
vided an excellent discussion of the rela- 
tions between these parameters. 

In commercial motion picture prac- 
tice, an over-all gamma from scene to 
screen of approximately 1.6 or 1.7 is con- 
sidered desirable from the audience 
point of view. It seems likely that the 
same objective also applies to television 
practice. In this case, however, the 
transfer characteristic of the kinescope, 
direct view or projection, is a power law 
with an exponent which probably falls 
in the range 2.0 to 2.8. Ideally, the 
transfer characteristic from scene illumi- 
nation to kinescope-grid driving signal 
should, in turn, be a power law with an 
exponent probably not exceeding 0.75. 
In order to compare camera-tube trans- 
fer characteristics, it will be necessary to 
assume that the line amplifiers are lin- 
ear. On this basis, the present studio 
practice of using a tube with a linear 
characteristic, such as the image orthi- 
con, results in raising the effective 
gamma above the desired objective, 
even in the case of live pickup where 
there is no modification of transfer 
characteristic due to film. 

Ordinary motion picture film can be 
assumed to have been processed to the 
over-all gamma figure mentioned above. 
When an iconoscope is used with such 
film at the illumination levels which are 
now common in film camera chains, the 



56 



January 1951 Journal of the S'MPTE Vol. 56 



result is an approximately linear trans- 
fer characteristic to the grid of the 
kinescope. In this situation, some 
transfer characteristic correction is 
probably desirable. When a tube with 
a linear characteristic is used in com- 
bination with the same type of film, the 
effective transfer characteristic to the 
grid of the kinescope has a power law 
exponent of about 1.6. The over-all 
transfer characteristic to the screen of 
the kinescope then has the extremely 
high power law exponent of 3.5 to 4.5. 
Hence, the use of some form of gamma 
correction is apparently mandatory 
when linear devices are used with 
normally processed film. 

As an alternative, film which has been 
specially proeessed for use with linear 
image pickup devices may be considered. 
However, film processing to an effective 
gamma of 1.0 is probably the minimum 
feasible. Such film would give, again, 
an approximately linear characteristic 
to the grid of the kinescope and would 
result in about the same effective over- 
all gamma as in the case of present 
studio cameras on live pickup. Such 
special processing is probably feasible 
for very large television stations, or for 
network operations, where the capital 
available and the magnitude of the 
operation may enable complete specifi- 
cation and control of all steps in film 
production. However, as a general 
approach to the problem of film camera 
chain design, it cannot be assumed that 
specially processed film will always be 
available. Any such design will there- 
fore have to include provision for 
gamma correction, and again must con- 
sider the effect of gamma correction on 
the system noise level. 

In any image tube, there is a noise 
level set by the fundamental photo- 
current associated with the first stage of 
the process. The noise current, /, 
for a given photocurrent, I p , in a band- 
width, A/, is given by: 



where e is the electronic charge = 1.59 X 
10 ~ 19 coulomb. 
For a 4.25-mc bandwidth, 

I np = 1.16 X 10~ 6 VT P amp. (2) 

Figure 5* illustrates this noise charac- 
teristic, which is typical of an ideal pick- 
up device and is approached by the 
photomultiplier. In the case of storage 
tubes, the noise level associated with 
other stages scanning beams, amplifier 
input circuits and the like masks the 
fundamental noise level almost com- 
pletely. This case will be discussed in 
more detail in connection with storage- 
type pickup tubes. 

Storage-Type Pickup Tubes 

Since the iconoscope, 3 the orthicon, 4 
the image iconoscope 5 and the image 
orthicon 6 ' 7 have been adequately de- 
scribed elsewhere in the literature, their 
construction and general mode of opera- 
tion need not be reviewed here. We 
may proceed directly to a consideration 
of those characteristics which are par- 
ticularly important for the film chain 
problem. 

Table I contains data from a number 
of sources, both published 7 - 8 and un- 
published, on camera tubes available at 
present. The Aeriscope and Photicon 
entries are based on manufacturers' 
information which has been supplied to 
the authors. Some obsolete tube types 
are included for comparison purposes. 
It may be noted that a very wide range 
of characteristics is tabulated. 

The smallest of the tubes is the Aeri- 
scope, an image iconoscope manufac- 
tured by Radio Industrie, in France, 
having a photosensitive area of ex- 
actly the same size as the 35-mm film 
frame. The Photicon, which is manu- 
factured by Pye, Ltd., of Cambridge, 
England, is also quite small, having an 
area less than one square inch. On the 
other hand, the mosaic area of the 



np = V2e/pA/ 



(1) 



* Similar to curves which may be found in 
Ref. 2. 



Garman and Lee: Film Camera Chains 



57 



X *" 

ll 






o ' 

CQ LJ*^ 



CO T-H O 

odd 



oooooo 



o o 



r^ S o 
co co O o O 

rH d d d o" 



o o o 

43 '43 '43 

I*|K| 

M O H O W 73 

2 g.2 S.8 O 



PH PH PH 



| & g 

a a * 5 5 a 

(O 03 53 M) <1) CO 

W w S W w w 



s g i 



i> i> oo co 

r-i -<& C3 O 

rH O O O 

d d d d 



o co 



oooooo 



00 OO 00 
<N <N <N 



lOlOCO rHrHCOCOCO 

C^l ** CO 00 OO O^ 0^ O) 

<N i-I d d d d d d o 



& a 



p - ; 
III 111 



c c 

^ o 

'5 3 

8 8 



January 1951 Journal of the SMPTE Vol. 56 



PHOTOCATHODE 
CURRENT 
IN 

MICROAMPERES 



IU 

in 2 


FOR Af = 


4.25mcps 






A 


/SIGN 


ALI 


iu 
in' 3 


AMPLIFIER 


WISE LEVEL 




/ 














/ 




R = IOO 




id 4 
id 5 


1300 




/] 


/ 




^ 


X NOIS 


EI NS 


MULTIPLIE 
GAIN 
REQUIRED 


R / 


^ 


R=IO 

i 


^ 






V 


^ 


A 




B 




u io' 7 ic> 6 id 5 ic 4 id 3 id' 2 10 

LIGHT FLUX ON PHOTOCATHOOE (LUMENS) (SENSITIVITY = 10/1 */ L ) 



Fig. 5. Noise characteristics for an ideal pickup tube. (From RCA Review*) 



1850A, the iconoscope most commonly 
used at present in film camera chains, is 
approximately 17 sq in. 

The sensitivities of the photo surfaces 
vary by a factor of approximately ten, 
the highest figures being obtained in the 
newest tubes, namely, the 5820 and 5826 
image orthicons and the European im- 
age iconoscopes. These same tubes 
also offer an advantage in that the spec- 
tral sensitivity curve of the photo sur- 
face very closely approximates the eye 
sensitivity curve, and hence enables 
very satisfactory operation with white 
light and color film. 

Tube sensitivity is normally specified 
in terms of the highlight illumination 
required on the photocathode. This 
form of specification is not very satis- 
factory for the television camera de- 
signer. The sensitivity of any photo- 
sensitive image device is more conven- 
iently measured in terms of total light 
flux required to give a picture with a 
specified signal-to-noise ratio, from a 
definite angular field of view, and with a 
specified depth of field. 9 Tube hand- 



books do not, of course, furnish informa- 
tion in this fashion, nor is there much 
indication of the signal-to-noise ratio 
attainable. In the case of pickup from 
film, depth of field is not an important 
criterion, but a knowledge of the total 
luminous flux required on the photo 
surface (independent of the picture 
size) is pertinent to any projector de- 
sign. The luminous flux required for 
each of these tube types is therefore 
tabulated as the product of the known 
area of the photosensitive surface and 
the nominal maximum highlight illumi- 
nation required. Approximate signal- 
to-noise ratios, in terms of peak-to-peak 
signal relative to rms noise voltage for a 
4.25-mc bandwidth are also tabulated. 
Maximum luminous flux and signal- 
to-noise ratio figures are illusory in one 
sense. For example, there is no strict 
limit on the illumination in the case of 
the iconoscope. Present practice, as a 
matter of fact, provides a highlight 
illumination of 40 to 75 ft-c on the 
mosaic of the 1850A in most film camera 
chains. On the other hand, the image 



Carman and Lee: Film Camera Chains 



59 



orthicon illumination is more or less 
limited to the values given. 8 

One entry in this table, labeled 
"Flashed Photicon" for want of a better 
term, refers to a particular method of 
operation of the Pye Photicon in film 
chains, rather than to a distinctly 
different type of tube. Details of the 
method are to be published elsewhere in 
the near future. However, the general 
features are known to the authors and 
are outlined herein by permission of R. 
Theile of Pye, Ltd., and F. H. Townsend 
of Cathodeon, Ltd., both in Cambridge, 
England, to whom acknowledgment for 
development is made. The timing 
diagram is basically similar to the third 
line of Fig. 2, except that flash illumina- 
tion as well as image illumination occurs 
during the vertical sweep retrace time. 
The flash illumination is provided by an 
auxiliary lamp which floods the photo- 
cathode with light during the initial por- 
tion of the retrace interval. The re- 
sulting photoelectrons provide a uni- 
formly distributed electron shower over 
the entire surface of the mosaic. Simul- 
taneously, the collector electrode is 
pulsed negative so that secondary 
emission is not collected from the mo- 
saic, which then becomes negative rela- 
tive to the normal collector voltage. 
The collector returns to normal poten- 
tial immediately following the light 
flash, and hence is appreciably positive 
with respect to the mosaic during image 
illumination and subsequent beam scan- 
ning. With a positive collector, sensi- 
tivity is increased and shading problems 
due to secondary electron redistribution 
are less acute. It is recognized that 
this kind of operation is possible only 
with intermittent exposure of the cam- 
era tube, as is the case in most film 
camera chains. 

Noise Considerations in Storage-Type 
Pickup Tubes 

The image orthicon represents the 
nearest approach to an ideal storage- 
type pickup tube. As was pointed out 



previously, the photocurrent noise is 
almost completely masked by the noise 
due to the scanning beam. The rela- 
tionship between scanning-beam-noise 
current, (7 n6 ), and scanning-beam cur- 
rent, (7 6 ), is identical with that given 
for the photocurrent noise and can be 
written as: 

7n6 = 1.19 X 10~ 6 A/76 (3) 

In the case of an ideal image orthicon, 
the beam current is 100% modulated. 
The beam-current noise is a maximum 
in the black, where the return beam cur- 
rent is a maximum, and a minimum in 
the whites, where the photocurrent 
noise is a maximum. This results in a 
total noise characteristic which is virtu- 
ally independent of illumination level, 
and in noise fluctuations which are 
approximately the same in the blacks 
and whites. Practically speaking, the 
image orthicon falls short of this per- 
formance because the efficiency of beam 
modulation is not greater than about 
25% or 30%. A very large part of the 
noise output from the tube is therefore 
due to the unmodulated beam noise. 
These relationships are illustrated in 
Fig. 6,* in which the inherent noise level 
for an ideal pickup device, the noise 
level for an ideal image orthicon, and the 
total noise actually obtained from an 
image orthicon are all plotted as a func- 
tion of light flux on the photocathode. 
It will be noted that the signal current 
and the inherent noise current are double 
the values shown in the previous figure, 
to account for the secondary emission 
multiplication of approximately 2, which 
occurs at the target. 

In the case of iconoscopes, orthicons 
and image iconoscopes which do not con- 
tain signal multipliers, the noise level is 
set by the associated amplifier noise 
level. The equivalent input noise, I nt , 
to an amplifier with a bandwidth, A/, 
and a response characteristic which is 



* Similar to curves which may be found in 
Ref. 2. 



60 



January 1951 Journal of the SMPTE Vol. 56 



flat and independent of frequency over 
that bandwidth is given by 10 : 



2 \^ 



1 + 



RtR(aC) 



! ) 



(4) 



where: k =* Boltzmann's Constant 
T = absolute temperature 
R = input resistance 



Rt = equivalent input resistance 



C 



due to shot noise in the first 

amplifier 

shunt capacity in the input 

circuit. 



This amounts to approximately 2.6 X 
10~ 9 amp for a flat 4.25-mc bandwidth. 
It is usually possible to increase sig- 
nal-to-noise ratio by using a fairly large 
load resistance. However, because of 
the associated capacity of the tube and 
input circuit, frequency compensation is 
required. It is necessary to peak the 
amplifier response characteristic to give 
a response which is proportional to fre- 
quency over the bandwidth. The 
equivalent input noise for a peaked- 
channel amplifier, (/'<), has been given 
by Schade 1 as: 

I'nt = 3.7 X 10- 19 (Af)V2 (5) 

which for a peaked 4.25-mc channel is 
approximately 3.4 X 10~ 9 amp. How- 
ever, although the measured noise cur- 
rent for this peaked channel is numeri- 
cally greater than that given for a flat- 
channel amplifier, the effect on the eye 
is actually less. This has been noted by 
Schade, 11 who has produced experimen- 
tal curves for the detail response charac- 
teristic of the human eye. He has 



shown that, because of the fine grain of 
the fluctuations associated with a 
peaked-channel amplifier, the effective 
noise current for a peaked channel is 
approximately one-third of the calcu- 
lated noise current for a flat 4.25-mc 
channel. The effective noise level hi 
such a channel is therefore reduced by 
the eye characteristic to approximately 
1.1 X 10~ 9 amp. 

Table II presents data on the transfer 
characteristic and the total luminous 
flux required for an effective signal-to- 
noise ratio of 35, for several of the tubes 
listed in Table I. The total luminous 
flux at maximum rating is listed for 
reference. On this basis, the icono- 
scopes and image iconoscopes offer a 
very much larger effective signal-to- 
noise ratio than is obtainable with either 
of the image orthicons, which have a 
much lower storage capacity and a flat- 
channel noise characteristic. 

Summary of Storage-Type Pickup 
Tube Characteristics 

The iconoscope can give a very good 
signal-to-noise ratio when used in a 
system having the proper transfer 
characteristic, but it presents difficulties 
with shading and bias lights, and does 
not have as good a spectral-response 
characteristic as might be desired. The 
image iconoscope is more sensitive, has 
fewer difficulties with shading, does not 
require edge or bias lighting, can give 
just as good a signal-to-noise ratio, and 
offers a very good spectral-response 



Table II. Transfer Characteristic and Effective Sensitivity of Television 

Camera Tubes. 



Tube Type 


Power Law Exponent 
of Transfer 
Characteristic 


Total Lumens at 
Max. Rating 


Total Lumens for 
Effective Signal-to- 
Noise Ratio of 35 


1850A 
1848 
1840 
Photicon 
Flashed Photicon 
5820 
5826 


0.7 
0.7 
1.0 (linear) 
0.7 
0.7 
1.0 (linear) 
1.0 (linear) 


1.17 
0.3 
0.11 

0.025 
0.01 
0.0001 
0.00043 


0.041 
0.03 
0.0098 
0.002 
0.001 
0.0001 
0.00012 



Garni au and Lee: Film Camera Chains 



61 



SIGNAL 

AND 

NOISE 

CURRENTS 

TO 

MULTIPLIER 
(MICROAMPERES) 



10" 

id 2 






FOR Af 


= 4.25 m 


cps 


<& 


>/ 






> 

id 3 

io' 4 


AMPLIFIES 


) NOISE LE\ 


rEL 






^ 


s 




R = 0.68 


7 






^ 


<iS 


INT 


[/TOTAL NOISE 




^ 






A/I NOISE 


- 


in 3 




^ ** 


.&f\ 

EFFECTIVE 








LIC 


HT RANGE 

3UMED 




10"* 
i 


/'" 


s 


AS 








LIGHT RANGE 




D; S io' 5 io- 4 io- s io- 2 * io- 1 i.o 

^PHOTOCATHODE CURRENT (MICROAMPERES)-^ 


ib- 7 io- 6 io- 3 io- 4 ib- 3 > io- 2 fcr 1 

ALIGHT FLUX IN LUMENS ON 10/t VL PHOTOCATHODE^ 



Fig. 6. Noise characteristics of image orthicons and iconoscopes. (From 

RCA Review 2 ) 



curve. The image orthicon is ex- 
tremely sensitive, requires no shading 
adjustments by the operator, and in the 
newer types has a good spectral-re- 
sponse curve; however, its signal-to- 
noise ratio is not very good when the 
transfer characteristic required for pick- 
up from film is considered. In terms of 
picture quality obtainable with storage- 
type camera tubes, the image iconoscope 
seems to rate first, the iconoscope second 
and the image orthicon third. How- 
ever, the image orthicon must not be 
discounted where low operating cost, 
rather than attainment of the very high- 
est picture quality, is of prime impor- 
tance. Ease of operation, and the rela- 
tively simple nature of the associated 
projection equipment are useful proper- 
ties for low-cost operation. It is quite 
feasible to consider a projector used on 
the studio floor with an ordinary studio 
image orthicon camera which is (Jollied 
up to the projector for film commercials 
and programs. There are, in fact, many 
kinds of film operations for which the 



image orthicon camera will give per- 
fectly acceptable picture quality, at 
low operating cost. 

Nonstorage-Type Pickup Tubes 

The earliest mechanical schemes of 
light-spot scanning, as applied to film 
pickup, utilized a rotating disc or drum 
as the source of the light spot and a 
photocell as the sensing, or transducing, 
element. The modern flying-spot scan- 
ner, using a special cathode-ray tube as 
the source of the light spot, and substi- 
tuting a photomultiplier for the diode 
photocell, is now widely recognized as a 
device which can provide very high 
quality signals from film. The multi- 
plier-type image dissector tube is also 
familiar to television engineers, and can 
produce excellent television pictures, 
but to date has received less publicity. 

By their nature, these nonstorage 
pickup devices require continuous illumi- 
nation of the photosensitive surface 
during the scanning of the picture. 
Hence, their use is confined either to 



62 



January 1951 Journal of the SMPTE Vol. 56 



continuous-motion projectors, or to 
those intermittent projectors in which 
film pulldown is completed during verti- 
cal retrace of the television scan. Un- 
fortunately, neither projector has as yet 
been successfully applied to pickup 
from motion picture film in this country. 

The necessity of scanning hi 2-3-2 
sequence, dictated by the difference be- 
tween television and motion picture 
frame rates in this country, very seri- 
ously complicates the problems of the 
continuous-motion projector. In Eng- 
land and Europe, where frame-for- 
frame playback is ordinarily used, the 
results obtained with a continuous- 
motion projector and the flying-spot- 
scanner technique are startlingly good. 

Where frame-for-frame playback is 
possible, the flying-spot-scanner tech- 
nique applied to pickup from continu- 
ously moving film offers some advan- 
tages over other methods of pickup. 
Adjustment of the centering, amplitude 
and linearity of the raster on the scanner 
tube allows compensation of certain 
types of imperfection in film motion. 
Experimentally, it has been found that 
the film velocity can be made suffi- 
ciently uniform to maintain good inter- 
lace and vertical resolution. Very high 
quality television pictures are obtained 
from film in this manner in equipment 
manufactured by the Cinema Television 
Co., and Electrical and Musical Indus- 
tries, Ltd., in Great Britain, and by 
Radio-Industrie hi Paris. 

The photomultiplier in this applica- 
tion constitutes a nearly ideal pickup 
device with a noise characteristic similar 
to that shown in Fig. 5. The highlight 
flux required for a very high signal-to- 
noise ratio is about 10~ 3 1m, which is not 
difficult to obtain from a high-voltage 
scanning tube especially designed for 
the purpose. 

The scanning tube presents more 
serious problems, such as the problem of 
phosphor "grain." Grain results in 
signal fluctuations which, on close in- 
spection, are seen to be nearly stationary 



on the raster. Experimental scanning 
tubes have been built which are rela- 
tively free of this defect, but such tubes 
are not as yet commercially available. 
Another problem is created by the phos- 
phor-decay tune of current tubes. 
Light output from the phosphor should 
decay to a low value in a fraction of a 
microsecond; otherwise light is col- 
lected from points along a line behind 
the flying spot, instead of from the spot 
alone, and streaking and loss of resolu- 
tion result. Although it is possible to 
compensate for slow phosphor decay by 
proper shaping of the frequency re- 
sponse curve of the amplifiers, the re- 
sults are not always optimum. Still 
another problem is due to phosphor 
color. The light output should be 
essentially white to enable faithful re- 
production of tonal values from color 
film. To date, the only phosphors 
found useful for flying-spot-scanner 
tubes have suffered the defect that the 
luminous spot is colored green, blue or 
violet. 

The principle of the image dissector 
tube is illustrated in Fig. 7. A steady 
and continuously illuminated picture is 
projected on the photocathode. By 
conventional television deflection tech- 
niques, the photoelectrons emitted from 
the photocathode are scanned across the 
stationary rear aperture and amplified 
by a more or less conventional electron 
multiplier. The projector may use 
either continuous film motion or rapid 
intermittent pulldown. The image dis- 
sector tube has the distinct advantage 
that there is no difficulty in rendition of 
color film, since a standard projector 
light source (tungsten or carbon) may 
be used. For the same reason, it is not 
difficult to obtain the light flux required 
(which is greater than that needed for 
the flying-spot scanner by a ratio equal 
to the number of picture elements 
scanned). 

A comparison of flying-spot-scanner 
and image-dissector techniques is diffi- 
cult because a consistent analysis must 



Carman and Lee: Film Camera Chains 



MAGNETIC FOCUSING COIL 



PHOTOCATHODE 




Fig. 7. Principle of 
the image dissector. 



-OUTPUT LEAD 



assume projectors developed beyond the 
point where they stand today. As a 
personal opinion, the authors submit 
that the continuous-motion projector 
will not be the solution; previous at- 
tempts do not seem to have yielded a 
steady enough picture for the dissector 
tube, and in the case of the flying-spot 
scanner the problem of conversion from 
24 to 30 frames/sec seems to be an in- 
surmountable obstacle. No intermit- 
tent projector capable of pulling film 
into register during the television field 
retrace time is now available. How- 
ever, in view of the many development 
groups at work on the problem, a satis- 
factory solution seems inevitable. Once 
such a projector is available, both flying- 
spot-scanner and image-dissector tech- 
niques will offer very interesting possi- 
bilities for generating high-quality tele- 
vision pictures from film. Both tech- 
niques promise ideal noise character- 
istic, complete freedom from shading 
problems, and relatively high sensitivity. 
At the present time, with projector 
mechanisms which are available, stor- 
age-type pickup tubes offer the only 
feasible solution. 

References 

1. O. H. Schade, "Electro-optical charac- 
teristics of camera systems" (Part III 
of a series on electro-optical character- 
istics of television systems), RCA Re- 
view, vol. 9, no. 3, pp. 490-530, Sept. 
1948. 



2. P. K. Weimer, "The image isocon," 
RCA Review, vol. 10, no. 3, pp. 366- 
386, Sept. 1949. 

3. V. K. Zworykin, G. A. Morton and L. 
E. Flory, "Theory and performance of 
the iconoscope," Proc. I.R.E., vol. 25, 
pp. 1071-1092, Aug. 1937. 

4. A. Rose and H. A. lams, "The orthi- 
con," RCA Review, vol. 4, no. 2, pp. 
189-199, Oct. 1939. 

5. H. lams, G. A. Morton and V. K. 
Zworykin, "The image iconoscope," 
Proc. I.R.E., vol. 27, pp. 541-547, 
Sept. 1939. 

6. A. Rose, P. K. Weimer and H. B. Law, 
"The image orthicon, a sensitive tele- 
vision pickup tube," Proc. I.R.E., vol. 
34, pp. 424-432, July 1946. 

7. R. B. Janes, R. E. Johnson and R. S. 
Moore, "Development and perform- 
ance of television camera tubes," RCA 
Review, vol. 10, no. 2, pp. 191-223, 
June 1949. 

8. RCA Tube Handbook. 

9. A. Rose, "A unified approach to the per- 
formance of photographic film, televi- 
sion pickup tubes and the human eye," 
Jour. SMPE, vol. 47, p. 273, Oct. 
1946. 

10. H. B. DeVore and H. lams, "Some 
factors affecting the choice of lenses 
for television cameras," Proc. I.R.E., 
vol. 28, pp. 369-374, Aug. 1940. 

11. O. H. Schade, "Characteristics of 
vision and visual systems" (Part I of a 
series on electro-optical character- 
istics of television systems), RCA 
Review, vol. 9, no. 1, pp. 13-37, Mar. 
1948. 



64 



January 1951 Journal of the SMPTE Vol. 56 



Characteristics of All-Glass 
Television Picture Bulbs 

By John L. Sheldon 



Discussed are methods of manufacturing glass television bulbs, together 
with engineering data on mechanical, dimensional, optical and electrical 
characteristics of bulbs and glass. Current trends are given for size, shape 
and deflection angle. 



AS AN ENGINEERING MATERIAL, glaSS 
has an extraordinary versatility 
and range of useful properties. The im- 
portant uses in motion pictures and tele- 
vision are too numerous to recite here, 
except to say that without glass it is 
difficult to see how the two industries 
could exist. In the case of television 
bulbs the properties of glass that are of 
particular importance are optical clar- 
ity, electrical characteristics and high- 
vacuum properties. 

Although picture-tube bulbs may be 
made of a combination of glass and 
metal, this paper will deal only with the 
all-glass type, which predominate in the 
industry today. Glass is basically a 
very cheap material and thus is a desir- 
able one on which to base a large volume 
item that must sell in a narrow-margin, 
competitive field. In addition to the 
fundamental price factor, glass is an 
electrical insulator and serves not only 
as the vacuum container, but permits 

Presented on October 16, 1950, at the 
Society's Convention at Lake Placid, 
N. Y., by John L. Sheldon, Development 
and Research Dept., Corning Glass Works, 
Corning, N. Y. 



the tube to be mounted in the set 
cheaply and with little danger of elec- 
trical leakage. 

Method of Manufacture 

Glass bulbs for small cathode-ray 
tubes, such as are used in oscilloscopes, 
are generally made by blowing in one 
piece. However, it is difficult to get the 
glass distribution and surface quality 
required for large bulbs by a blowing 
method. Therefore present-day large 
bulbs are made by a process which was 
originated by Corning Glass Works, 1 the 
first large-scale application being the 
production of large quantities of cath- 
ode-ray bulbs used during the war in 
radar equipment. 

Briefly, the present method of manu- 
facture consists in sealing together three 
separate parts. In Fig. 1 are shown the 
parts from which the popular 16-J^-in. 
rectangular bulb are made. The panel 
is made by pressing, which insures accu- 
rate control of face thickness and curva- 
ture. The middle section, or funnel, 
also may be made by pressing, or by a 
process of centrifugal casting. Cast 
funnels have the advantage of less 



January 1951 Journal of the SMPTE Vol. 56 



65 



weight. Drawn tubing is used for the 
neck, thus satisfying the rather stringent 
requirements imposed by close-fitting 
components that must slide over the 
neck, as well as the need for ample elec- 
tron beam clearance inside. Further, an 
accurate, round bore insures accurate 
alignment of the electron gun. 

A #4 Alloy "button" is sealed into the 
side of the funnel with fully automatic 
machinery. It serves to make contact 
with the conductive coating which is on 
the inside of the bulb. Successful but- 
ton sealing goes back to the manufacture 
of the alloy. The analysis must be 
within close limits, as well as the expan- 
sion coefficient. Also, it must have 
proper oxidation characteristics. Be- 
fore use, the button must have a special 
cleaning, followed by oxidation in wet 
hydrogen at about 1200 C. The oxide 
that is produced bonds to the glass 
during sealing to form a strong, vacuum- 
tight joint. 

Formerly the three separate parts 
were joined by welding with gas fires. 
We now employ an electric method 2 for 



sealing panel and funnel together. It 
has the advantage that heat is gener- 
ated within the glass, rather than 
"pushed" in from the outside. Also, the 
method is fast and easy to control, 
hence it is very well suited for mass 
manufacture. The result of electric 
heating is a seal that has excellent 
geometry and strength. 

Because the finished tube is evacu- 
ated, and thus subject to external pres- 
sure, it must be strong. A factor of 
safety is necessary to guard against 
breakage when the tube is mishandled. 
Therefore it is customary to design 
bulbs to withstand a pressure of three 
atmospheres. 

Glass Characteristics 

One of the important developments 
of the past two years has been that of a 
new lead-free glass designed particularly 
for mass-produced picture bulbs. Dur- 
ing the war lead glass was used for 
radar cathode-ray oscilloscope bulbs, 
partly because high electrical resistivity 
was needed. Lead has been an expen- 




66 



Fig. 1. Separate parts that are sealed together to make a bulb. 
January 1951 Journal of the SMPTE Vol. 56 



sive and uncertain material and because 
a substantial percentage was used, the 
bulbs were heavy. 

A glass completely free from lead is 
of particular importance at this time 
when it is almost certain that restric- 
tions will be placed on lead and other 
strategic materials. This is doubly im- 
portant in view of the accelerated elec- 
tronics program. Radar tubes can be 
made from lead-free glass to advantage; 
in fact, the optical quality of future 
radar tubes will be far better than those 
used in the last war. Military radar 
will also benefit from other substantial 
advances made in glass technology. 
During World War II most of the panels 
for radar bulbs were made by the labor- 
ious method of hand pressing. Now, 
high-speed pressing of 20-in. panels is 
routine. 

The new glass (Corning Code 9010) 
was tailored to the exacting require- 
ments of television. It has a high elec- 
trical resistivity, is 15% lighter than the 
lead glass formerly used (Corning Code 
0120) and can be readily melted to give 
the exactingly high quality that is de- 
manded in picture- tube panels. In 
Table I are some engineering data. 

Table I 

Density 2.59 

Refractive Index (N D ) 1 . 506 

Coefficient of Expan- 
sion (Average 0- 

300 C) 88.5 X 10~ 7 

cm/cm/ C 

Electrical Resistivity 

350 C log 7.0 (ohms/cm) 

250 C log 8. 9 (ohms/cm) 

Softening Point . . 650 C 

Annealing Point . . 442 C 



Strain Point . . . 



411 C 



Although it is important to control 
the properties of all electronics glass 
within narrow limits, this is particularly 
vital for television picture bulb glass. 
Close control of the expansion coefficient 
is dictated by the method of bulb manu- 



facture, which requires very large seals 
between relatively thick "high" expan- 
sion glass. In this operation, two glass- 
to-glass and one glass-to-metal seals are 
required. Much of the time the sepa- 
rate parts are produced from different 
tanks. A third glass-to-glass seal is 
made by the tube manufacturer. Not 
only is expansion important, but so also 
are the viscosity characteristics. The 
"stem" carrying the electron gun is 
joined to the neck with a "drop" seal, 
in which the heated neck-glass is pulled 
down around the stem by gravity. This 
high-speed, automatic operation is de- 
pendent on close control of glass prop- 
erties. 

High electrical resistivity is desirable 
for several reasons. First, the full 
anode voltage appears across the wall of 
the neck tubing, from the inside conduc- 
tive coating to the external components 
which are at ground. Second, in many 
types the outside of the funnel is coated 
with a conductive paint. Thus, the 
bulb also serves as a filter condenser, the 
glass wall being the dielectric. Last, if 
the resistance were low, then the tube 
mounting would have to be a good insu- 
lator to prevent excessive electrical leak- 
age through the glass to ground. This 
would increase set cost. 



I 


\ 














/ 


/***" 


i- 


> 


\ 












/ 






v_ 


s 


-\ 


s~\ 


' Ss - 


-J 







420 



500 



740 



580 
Xlw/0 

Fig. 2. Transmittance curve for 
Corning 9010 neutral-gray glass. 

The first lead-free glass to be used was 
"clear." However, following an exten- 
sive program of development with the 
tube industry, a neutral gray version 
was offered in 1949. The spectral trans- 
mittance is shown in Fig. 2. Use of a 



John L. Sheldon: All-Glass Picture Bulbs 



67 



neutral absorbing glass minimizes the 
loss of contrast due to ambient light 
that falls on the screen. 3 This was an 
important contribution, because of the 
trend toward viewing television in 
lighted rooms and, also, because of the 
increase in the number of daytime 
programs. An absorbing glass also 
minimizes loss of contrast due to hala- 
tion, which is the result of internal reflec- 
tions within the face. 4 

At present there is an industry stand- 
ard for luminous transmittance and 
chromaticity which was agreed upon by 
the Joint Electron Tube Engineering 



Committee (JETEC) of the RTMA 
(Radio and Television Manufacturers 
Assn.). For 10^-in. and 12^-in. bulbs 
the luminous transmittance is 66 3%. 
The chromaticity is denned by use of the 
International Commission on Illumina- 
tion color system. In Fig. 3 is shown a 
nominal spectral emission curve for the 
P4 7000 white phosphor used in tele- 
vision tubes, while Fig. 4 shows the 
tolerance area for chromaticity. The 
nominal chromaticity of the standard 
P4 phosphor-gray glass combination is 
x = 0.3044 and y = 0.3177, with a toler- 
ance area as shown in Fig. 5. Considera- 



400 440 480 520 560 600 640 680 720 




Fig. 3. Spectral energy emission characteristics of typical 7000 K all-sulfide 

P4 screen. 



January 1951 Journal of the SMPTE Vol. 56 



tion is currently being given to stand- 
ardizing larger-sized bulbs. 

Trends in Bulb Design 

The phenomenal growth of television 
is matched only by the equally rapid 
rate of change within the art a rate 
so great as to make most written mate- 
rial out-of-date before it can be pub- 
lished. 

1. Size. In 1948 the 7-in. electro- 
static and 10^-in. electromagnetic 
round tubes were the large-volume 
types. They were supplanted by the 
12^2-in. round tube in 1949, which in 



turn has become practically obsolete, 
being followed by 16-in. and 19-in. 
round bulbs. This evolution is shown 
in Fig. 6. 

The ready acceptance of larger and 
larger pictures brought about the prac- 
tice of "overscanning," which resulted 
in a picture with straight top and bot- 
tom sides, but with circular ends. While 
this increased the utilization of avail- 
able screen area, there was a loss of the 
information in the corners and a depar- 
ture from the 4:3 rectangular shape. 
This subject was recently discussed by 
Bretz. 5 



.370 



.360 



.350 



.340 



.320 



.310 



.300 



.290 




7000K 



/8000K > 




'10000 K 



80 .290 .300 .310 .320 .330 .340 .350 

X 

Fig. 4. JETEC color limits for P4 white phosphor. 
John L. Sheldon: All-Glass Picture Bulbs 



69 



.3200 



.3150 



.3100 




.3000 



.3050 



3100 



Fig. 5. JETEC specifications for neutral filter face glass. 




Fig. 6. Trend of bulb size, 1948-1950. 

From left to right, 7-in., 8^-in. (both blown bulbs), 10^-in., 12^-in., 15%-in., 



70 



January 1951 Journal of the SMPTE Vol. 56 







Fig. 7. Three 15 %-in. bulbs 
Left to right, 70, 52 and 60 deflection angles. 




Fig. 8. Rectangular bulbs 

Left to right, IS^e-in., 16i-in., 20%2-in. 



John L. Sheldon: All-Glass Picture Bulbs 



71 





Major Axis 



Minor Axis 



23" R. 




Panel 



Diagonal 

Fig. 9. Drawings of 16^-in. rectangular bulb. 



2. Length. While round bulbs were 
still in use there was a trend toward 
wider deflection angles, which resulted 
in the desirable advantage of shorter 
tubes. For example, the 15K-in. round 
all-glass bulb has been made in 52, 60 
and 70 types, as shown in Fig. 7. All 
use the same panel, but different fun- 
nels. Shortening the bulb saves valu- 
able cabinet space and this has become 
more important with the trend to larger 
sizes. At the time of writing, 70 is the 
commonly used deflection angle. 



3. Bulb Shape. The somewhat dubi- 
ous practice of overscanning in round 
tubes has now been corrected through 
the introduction of rectangular bulbs. 
We are sure that the return to the rec- 
tangular picture is gratifying to most of 
the members of this Society. Figure 8 
shows the I3 1 ^{ 6 -m., 16%-in. and 
20%2-in. bulbs. At the time of writing, 
the 16^-in. is a very popular type, 
although the demand for 20% 2 -in. is 
increasing rapidly. 

4. Dimensional and Other Considera- 



72 



January 1951 Journal of the SMPTE Vol. 56 



tions. The rapid rate of change has 
brought with it an engineering challenge 
of some magnitude and the problems of 
designing and building equipment of 
increasing size have resulted in consider- 
able progress in the art of glass making. 
The demand for better and better glass 
quality has led to frequent revision of 
specifications and we should not here 
attempt to go into the six or seven pages 
of specifications that cover a single type, 
except to say that the glass quality and 
dimensional standards have steadily in- 
creased. Some of the conventions hav- 
ing to do with dimensions may be of 
interest. In Fig. 9 are outline drawings 
of the 16^-in. rectangular bulb, which 
show some of the important dimensions. 
A practice of long standing in the lamp 
and tube industry is to rate the size of 
bulbs by use of a number which is the 
maximum outside diameter in eighths of 
an inch. Rectangular bulbs are rated 
by the diagonal dimension. For ex- 
ample, the 16%-in. bulb shown in Fig. 
9 is a C-133. When the bulb is regis- 
tered with the American Standards 
Assn. the size designation is prefixed 
with the letter "J." Tube sizes are also 
based on the maximum outside diam- 
eter, or diagonal, and are given in 
inches, to the nearest inch. Tubes are 
registered with the Radio and Television 
Manufacturers Assn., which assigns a 
title. For example, one of the tubes 
made from the 16^-in. bulb is the 
17AP4. The "A" is a serial designation 
and the P4 describes the phosphor. 

Future Developments 

We believe the rectangular shape is 
here to stay. The 20% 2 - m - size is be- 
coming very popular, but it does not 
appear to be the end and a rectangular 
bulb with a diagonal in the mid-20's is 
on the drawing board. The ultimate 
size will probably be limited by eco- 
nomic considerations and certainly by 
the size of the average door. How much 
wider the deflection angle will go is a 



matter that depends more upon circuitry 
and component considerations than on 
glass manufacturing. However, the 
larger sizes will bring pressure to 
shorten the bulb through use of wider 
angles. 

To date virtually all the tubes manu- 
factured have gone into new sets. It is 
the writer's opinion that there is a place 
in the home for a medium-sized picture, 
perhaps in the 14-in. range, and that 
in the future there will be a return to 
this size. Such a set might well be the 
"second" one in the home. 

As of October, 1950, there is consid- 
erable interest in the use of nonglare 
finish on the face of tubes. This is a 
slight matte finish which diffuses reflec- 
tions and thus lessens the annoyance 
due to recognition of various objects or 
light sources that are seen by specular 
reflection in the untreated tubes. Such 
a finish must be carefully controlled to 
strike the best compromise between 
reduction of specular reflection and loss 
of resolution and contrast. 

A more recent solution to the problem 
of annoying specular reflections is the use 
of panels that have a cylindrical, rather 
than spherical surface, the axis of the 
cylinder being vertical. It is obvious, 
from simple geometry, that in most cases 
the seated viewer will not see reflections 
of lights and objects that are above his 
eye level and, also, that further protec- 
tion can be realized by tilting the tube 
downward a few degrees. As a result, 
the room can be easily lighted to a de- 
sirable level without the annoyance of 
reflections. Also, there is no loss of 
resolution or contrast as is the case with 
tubes having a frosted finish. Demon- 
strations of operating tubes were held 
for tube and set makers in New York 
and Chicago in late October and the re- 
sults were very striking. 



Acknowledgment. A. E. Martin, Syl- 
vania Electrical Products, Inc., very 
kindly furnished illustrations used in 
Figs. 3, 4 and 5. 



John L. Sheldon: All-Glass Picture Bulbs 



73 



References 

1. Huston Harris, U.S. Pat. 2,160,434. 

2. E. M. Guyer, "High frequency electric 
glass welding," Trans. Electrochem. 
Soc., vol. 79, pp. 187-198, 1941; "Elec- 
tronic welding of glass," Electronics, 
vol. 18, pp. 92-96, 1945. 

3. Alfred E. Martin and Robert M. 
Bowie, "Picture-tube contrast im- 
provement," Electronics, vol. 23, pp. 
110-112, Aug. 1950. 

4 R. R. Law, "Contrast in kinescopes," 
Proc. I.R.E., vol. 27, pp. 511-524, 1939. 

5. Rudy Bretz, "The shape of the tele- 
vision screen," Jour. SMPTE, vol. 54, 
pp. 545-553, May 1950. 

Discussion 

MR. SEELEY: Would the author say a 
few words about glass tubes as compared 
with metallic tubes, with regard to the 
results that can be obtained and the cost of 
production? 

DR. SHELDON: I am not prepared to dis- 



cuss the production costs. The perform- 
ance is a matter of tubes, and I think 
that is a question that might more prop- 
erly be answered by one of the tube people. 
So far as I know, there is no essential 
difference in the performance, once the 
tube is installed in the set. However, 
all-glass tubes have certain advantages as 
regards mounting in the set. 

ANONYMOUS: What is the minimum 
reflection on the face of a 20-in. tube the 
minimum arc across the surface? 

DR. SHELDON: You mean the panel 
radius? 

ANONYMOUS: That is right. 

DR. SHELDON: The outside panel radius 
is 40 in. 

ANONYMOUS: What is the chord across 
the curved surface of the face of the tube? 

DR. SHELDON: Across the maximum 
diagonal? That is about 1%-m. less than 
that (20 in.), approximately. That takes 
care of the thickness of the glass and the 
radius. 



January 1951 Journal of the SMPTE Vol. 56 



ABSTRACT 



Stereo-Television in Remote Control 



By H. R, Johnston, C. A. Hermanson, and H. L. Hull 



THE STUDY of the possibilities of 
using three-dimensional television 
in conjunction with remotely controlled 
electric manipulators is part of a long- 
range development program being 
undertaken by the Remote Control 
Engineering Division of the Argonne 
National Laboratory. 

Manipulation of objects in three di- 
mensional space requires that depth 
perception be incorporated into any 
scheme used to view and control the 
means of manipulation. It is not suffi- 
cient to use ordinary two-dimensional 
television for this purpose since the 
ability to judge depth is almost entirely 
lacking. 

A standard Du Mont television pick- 
up chain was employed in the develop- 
ment of stereo- television. The stereo- 
scopic pair of images are placed side by 
side by a twin lens system onto the 
photocathode of the television camera 
tube. The images occupy the same 
space on the photocathode as a single 
image in standard two-dimensional tele- 
vision and they are transmitted simul- 
taneously. At the receiving end of the 
stereo-television system, the two images 



Abstract by Pierre Mertz of a paper pre- 
sented on September 26, 1950, at the 
National Electronics Conference at Chi- 
cago, 111., (in which the SMPTE Central 
Section participated), by H. R. Johnston, 
C. A. Hermanson and H. L. Hull, Argonne 
National Laboratory, P.O. Box 5207, Chi- 
cago 80. The complete paper was pub- 
lished in Electrical Engineering for Decem- 
ber, 1950, and will also be published in 
Proceedings of the National Electronics Con- 
ference, vol. 6 (for 1950). 



appear side by side on the face of a 
standard kinescope or television picture 
tube. 

Two polarizing filters whose axes of 
polarization are at right angles to each 
other are placed immediately in front 
of the images on the cathode-ray tube. 
An observer wears a pair of polarizing 
spectacles so oriented that the right eye 
is permitted to see only the right-eye 
image and the left eye sees only the left- 
eye image. 

A second method used to view the 
three-dimensional television pictures 
makes use of two television picture 
tubes. These tubes are arranged at 
right angles to each other and a semi- 
transparent mirror is placed so that it is 
at 45 with both tubes. Crossed polariz- 
ing filters are placed in front of each pic- 
ture tube and the observer wears crossed 
polarizing spectacles. The observer is 
enabled to see the three-dimensional 
image by observing one image by trans- 
mission through the semitransparent 
mirror and the second image by reflec- 
tion. 

To test adequately the possibilities of 
the stereo-television system as a means 
of seeing objects in three-dimensional 
space, two mechanical Master-Slave 
manipulators were arranged so that the 
operator sat with his back to a wall, 
behind which the slave hands and the 
stereo-television were located. The 
operator faced the stereo receiver and 
saw a three-dimensional image of the 
manipulator "slave" hands and objects 
in the work area, while with his hands in 
the "master" controls he manipulated 
objects in the field of view. After a few 



January 1951 Journal of the SMPTE Vol. 56 



75 



minutes of indoctrination any person 
with normal vision can be taught to see 
and manipulate the objects in view from 
a remote distance. In another setup, 
an electrically operated manipulator was 
made to perform miscellaneous feats of 
lifting objects and pouring liquids from 
one beaker to another, while the opera- 
tor controlled its movements from an- 
other room over 50 ft away. 

The present system of stereo-televi- 
sion using one camera pickup tube, gives 
a stereo picture which has an aspect 
ratio of three high and two wide. This 



may be undesirable for use in any per- 
manent installation. In addition, the 
field of view is restricted, and the resolu- 
tion is adversely affected. 

A more desirable system would con- 
sist of the use of two television camera 
pickup tubes arranged side by side in a 
horizontal direction. The left pickup 
tube would supply a left-eye view to one 
of the receiving tubes of the dual viewer 
and the right pickup tube would supply 
the video signal for the second receiving 
tube. 



ABSTRACT 



The Orthogam Amplifier 



By C. L. TWnsend and E. D. Goodale 



TT1OR SOME TIME it has been known that 
f iconoscope film pickup tubes 1 - 2 do 
not produce video voltages ideally suited 
for reproduction by a normal kinescope 
unless gradient correction is applied. 
A re-evaluation of the transfer charac- 
teristic required in the television trans- 
mission system for optimum picture 
quality was undertaken, to include 
conditions actually encountered in nor- 
mal commercial broadcast operation. 

A series of slides was produced, each 
having an "average gray" background 
(density about 1 .2) and, centered in that 
area, a rectangular "window." Each 
slide was made with a different window 
density, to cover the range normally 

Abstract by Clyde R. Keith of a paper by 
C. L. Townsend and E. D. Goodale, Engi- 
neering Dept., National Broadcasting Co., 
Inc., RCA Bldg., Radio City, New York 
20, published in RCA Review, vol. 9, no. 3, 
pp. 399-410, Sept. 1950. 



encountered in practice. The slides 
were projected in succession, and in 
such a way that the same portion of the 
mosaic was used for each window. Os- 
cilloscope readings of the voltages so 
generated showed a reasonably linear 
relationship with window density. 

To determine the characteristic which 
is actually obtained in the existing re- 
cording-reproducing system, the "win- 
dow" test was again used. A video 
voltage, representing the window and an 
appropriate background, was fed to the 
recording system. The amplitude of the 
voltage of the window proper could be 
precisely controlled to produce any value 
within the normal recording range, plus 
some excess into overload values, if de- 
sired. A recording was made of this 
signal, and the film processed normally. 
That film was then reproduced on an 
iconoscope system, and the output volt- 
age values noted on an oscilloscope. 



76 



January 1951 Journal of the SMPTE Vol. 56 



i 



The resulting plot is shown in Fig. A 
[Fig. 3 in original paper], and includes a 
wider range than is normally used. A 
serious compression of white-range volt- 
ages is present. All transfer characteris- 
tics of the intermediate recording and 
reproduction steps also were plotted 
with significant information produced at 
each point. Exposure and processing 
methods were altered in an effort to re- 
duce the undesirable effects shown in 
Fig. A, but the general characteristic re- 
mained. 

An analysis of Fig. A indicates that 
some nonlinear compensation is needed 
for both direct film and kinescope re- 
cording reproduction. The two cases 
differ as to amount required, but are 
otherwise generally similar. Both are 
simple curves, and compensation should 
be feasible. 

Many times in the past "gamma cor- 
rection" amplifiers have been built 
which had variously shaped transfer 
characteristics. Most of these actually 
compressed one part of the characteris- 
tic in order to get a relative expansion 
of another. Figure A indicates that the 
"black" half of the characteristic should 



not be altered, but rather an expansion 
in the "white" range is required. No 
gradient change should be permitted in 
the near-black signals, even though their 
relative amplitude is reduced to permit 
white expansion. 

With the above requirements in mind, 
the Model "A" orthogam amplifier was 
designed with two parallel amplifiers, 
as indicated in Fig. B [Fig. 4 in original 



Fig A. Volts input to kinescope re- 
cording versus volts output from film 
reproducer. 



LINEAR AMP 



LINE 

OUTPUT 




Fig. B. Block diagram of Model "B" orthogam amplifier. 
Townsend and Goodale: Orthogam Amplifier (Abstract) 



77 



paper]. The upper branch path pro- 
vides a completely linear output voltage, 
to which the lower branch adds ex- 
panded white voltages. Large video 
voltages can be fed to V5, and its bias 
can be controlled to allow only the high- 
light tips of those voltages to be passed 
by the tube. Thus both amount and 
gradient of the correction can be con- 
trolled, without causing nonlinear opera- 
tion in the black region. 

Six orthogam Model "A" amplifiers 
were put into operational service as an 
extended performance check. "In-Out" 
tests immediately gained the coopera- 
tion of the operating personnel. Notice- 
able improvement in film transmission 
quality was commented on by observers 
not familiar with the tests. As an un- 
expected dividend in many cases the 
effect of flare was reduced, since in 
normal operation most of it occurs at 
low amplitude in dark areas, and the 
orthogam reduces the relative amplitude 
of such voltages. However, it shortly 
became apparent that some changes in 
the method were required. Operating 
crews found the units were "wild" that 
is, they made video level riding difficult. 
This was found to be due to the fact that 
once a correct gradient was chosen for 
the normal maximum voltage level, 
much steeper gradients existed above 
that point, in the nominally unused 
region of overload. Frequently a video 
voltage peak would rise into that re- 
gion and the additional amplification 
there would drive it far higher than it 
would otherwise have gone. Subse- 
quent reduction of system gain cor- 
rected the matter, but only after a 
troublesome transition period. With 
close attention during rehearsal, and 
constant vigilance during broadcast, 
these effects could be acceptably mini- 
mized, but final judgment was that the 



Model "A" was not an operationally de- 
sirable tool. 

Based on the above results, a new 
attack on the problem was made. Us- 
ing the same basic philosophy of cor- 
rection, it was decided that the major 
additional requirement was that the top 
desired gradient must be the greatest 
actually encountered in the system 
under operating conditions. Thus, 
instead of a continuously rising gradient 
in the overload-voltage region, the new 
orthogam must be linear at the steepest 
desired slope. This objective has been 
achieved in the Model "B" orthogam 
amplifier. [A circuit schematic and 
description are given in the original 
paper. ] 

Several NBC film studios now have 
been equipped with Model "B" units, 
and considerable operational experience 
indicates that the gainriding difficulty 
experienced with the "A" model has 
been largely overcome and substantial 
improvement provided in the transfer 
characteristic of the over-all system. 
This is evidenced in the viewed picture 
by a reduction in the chalkiness of faces 
and an improvement in the separation 
between other white and near-white por- 
tions of the reproduced image. The 
average brightness of the picture is re- 
duced somewhat due to the fact that the 
a-c axis has been pushed towards the 
blacks. The end result is a more nat- 
ural and pleasing reproduction. 

References 

1. O. H. Schade, "Electro-optical charac- 
teristics of television systems," RCA 
Rev., vol. 9, nos. 1, 2, 3, 4, Mar., June, 
Sept., Dec., 1948. 

2. D. G. Fink, "Brightness distortion in 
television," Proc. I.R.E., vol. 29, no. 6, 
June 1941. 



78 



January 1951 Journal of the SMPTE Vol. 56 



Diffuse and Collimated T-Numbers 

A Review and Description of New Equipment 
By Allen E. Murray 



The SMPTE Subcommittee on Lens Calibration has formally recognized, 
through incorporation in its report, two methods of lens calibration. 
While they reach equivalent results and will calibrate lenses identically 
when properly safeguarded, each has its own shortcomings and advantages, 
which are not commonly recognized. To dispel the evident misunder- 
standings about these two methods, they are compared and the reasons are 
indicated for the method chosen. New equipment designed by Bausch & 
Lomb Optical Co. for lens calibration based on the diffuse method is de- 
scribed briefly. 



FROM THE INTERDEPENDENCE of 
physical phenomena it follows that 
a given quantity can be measured in 
more than one way. The more funda- 
mental it is, the larger is the number of 
its interrelationships and the larger 
the number of methods available for its 
evaluation. In the realm of pure phys- 
ics this principle is put to use in assur- 
ing consistency of theories and the 
correctness as well as the limits of 
accuracy of fundamental constants. 
On the engineering level it assures, in 
addition, the solution of virtually 
any problem that may be raised, since 
it provides alternative procedures for 
finding the solutions, as well as checks 
on their correctness. 



Presented on October 19, 1950, at the So- 
ciety's Convention at Lake Placid, N.Y., 
by Allen E. Murray, Scientific Bureau, 
Bausch & Lomb Optical Co., Rochester 2, 
N.Y. 



It is interesting to note that there is 
also a complementary principle at work : 
the principle of uniqueness of experi- 
mental arrangements and implications. 
This requires that the results obtained 
from given experimental equipment be 
specific to that equipment and issue 
only from the principles it employs. 
The distinctions thereby created may 
in many cases be without a difference, 
but the possibility of alternative 
methods of accomplishing things im- 
plies that the things done are not 
identical in detail. 

These observations are prompted by 
the recent history of photometric lens 
calibration. This problem reduces, in 
essence, to the measurement of the 
illuminance in the image plane of a 
lens under such conditions that the 
transmittance and relative aperture 
are evaluated together. This requires, 
in principle, that the illuminance at a 
particular stop be compared with the 



January 1951 Journal of the SMPTE Vol. 56 



79 



illuminance produced by an ideal lens 
or its equivalent at a similar stop. 
Within the limitations of the general 
method there is a large number of 
procedures capable of meeting the 
requirements of accuracy and obedience 
to essential physical principles. These 
procedures are not all equally reliable 
or sound, and each measures the illumi- 
nance under a different set of circum- 
stances, i.e,, measures a quantity 
which, though related, is not always 
the one wanted. 

All these considerations were in the 
collective mind of the Society's Sub- 
committee when it prepared its report 1 
and, in Appendix II, discussed two 
general experimental procedures 
whereby the T-number could be evalu- 
ated. The Subcommittee, however, 
understandably failed to point out that 
of all physical procedures, the photo- 
metric are among the most treacherous, 
in that even when the principles are 
correctly applied, it is disconcertingly 
easy to commit some simple error of 
omission, failure to eliminate every last 
trace of stray light for instance, vitiating 
the whole procedure. The extent to 
which extremely careful attention must 
be paid to every safeguard in photomet- 
ric practice is not realized by those un- 
familiar with photometry, and as a re- 
sult many proposals of unequal merit 
have been published from time to time. 

The Subcommittee also failed, and 
equally understandably, to point out 
that the two methods, being different ab 
initio, must evaluate different physical 
quantities, and moreover each must 
have its own set of shortcomings and ad- 
vantages. 

The Subcommittee was following his- 
torical precedent when it chose to de- 
scribe the collimated and the diffuse 
source methods, for the published pro- 
cedures have fallen naturally into the 
same two classes. These published 
methods are included here for their his- 
torical interest, and are further classi- 
fied according to whether the light is 



sent in the normal or counter direction 
through the lens: 

/. Collimated Source 

Normal Counter 

Silvertooth 2 Odencrants 6 

Daily" Hrdlicka 6 

Townsley 4 

//. Diffuse Source 

Normal 



Counter 
Berlant 14 
Murray 16 



Lambert 7 

McRae 8 

Moffitt 9 

Clarke &Laube 10 

Sachtleben 11 

Gardner 12 

Back 13 



A balance sheet of the several advan- 
tages and disadvantages in principle and 
practice can be drawn up without dif- 
ficulty. After noting that the colli- 
mated methods in effect evaluate a 
quantity proportional to the diameter 
of the entrance pupil of the objective, 
while the diffuse methods evaluate the 
flux on the image side of the lens, an 
unimportant distinction for most pur- 
poses, the two methods can be compared 
on their merits as procedures yielding 
the T-number defined by the Subcom- 
mittee. 

/. Collimated Source 

Advantages 

1. Focusing unnecessary 

2. Lens always correctly focused 

3. Little power required in source 

Shortcomings 

1 . Knowledge of equivalent focal length of 
lens essential 

2. Requires different set of apertures for 
each focal length or calibrating 
means such as Townsley's 4 

3. Theory more complex 

4. Indirect measurement of T-number 

5. Uniformity of collimated beam is 
troublesome to ensure; the effect of 
beam spread is difficult to evaluate 

6. Not directly adaptable to finite magni- 
fications 

7. Entrance pupil diameter limited by 
collimator lens 



January 1951 Journal of the SMPTE Vol. 56 



//. Diffuse Source 

Advantages 

1. Focal length knowledge unnecessary 

2. Adaptable to any magnification (with 
focal length known ) 

3. More fundamental and thus simpler in 
principle 

4. Maximum lens aperture unlimited 

Shortcomings 

1. Focusing essential 

2. Attainment of uniform source quite dif- 

ficult 

3. Light losses large high sensitivity in 
detector or great power in source re- 
quired 

The criticisms are readily seen to be 
unequal in weight, numbers 3 and 4 
under the collimated source being minor 
objections in the theory, while 2 and 5 
can be overcome by careful engineering. 
The most serious of these shortcomings 
are perhaps 6 and 7, and in this order. 
It is no trick to measure the equivalent 
focal length accurately enough (1), but 
the limitations to lenses of a given di- 
ameter and always at a fixed magnifica- 
tion are real handicaps. The colli- 
mated source method demands that the 
collimator lens always be larger than the 
entrance pupil of the lens being tested, 
and this requires costly lenses in larger 
sizes. The author knows of no simple 
way of adapting this method to finite 
magnifications. 

The advantages of this method are all 
substantial: to have the lens under test 
automatically and securely focused un- 
deniably creates confidence, and the 
convenience of a light source whose 
power requirements are small is not to 
be denied. 

Except for the first, the shortcomings 
of the diffuse method are serious enough 
to demand the most careful engineering. 
Focusing is easy; it can be done by au- 
tocollimation or by the use of a tele- 
scope. It demands considerable engi- 
neering effort, however, to ensure an ex- 
tensive diffuse source whose uniformity 
is sufficient to meet the requirements, 



and in addition sound design to attain 
useful sensitivity with reasonable power 
input into the lamphouse. 

Numerous methods of assuring uni- 
formity with diffusion have been pro- 
posed. Perhaps the best is the one 
proposed in the Subcommittee Report, 
using a sheet of direct-light shielded 
ground glass to cover the aperture in a 
matte white box. Even at best, how- 
ever, these lamphouses must be large, 
since it is necessary that the T-stop 
equivalent solid angle be filled with flux 
at all values. Fortunately the measure- 
ments are independent of the source dis- 
tance when the incident cone is filled. 

The advantages of this method coun- 
terbalance the disadvantages. It is 
clear that adaptability to all magnifica- 
tions and no restrictions on lens aper- 
ture together make up a strong argu- 
ment in its favor. 

These considerations seemed to us to 
be so cogent that when we designed the 
equipment to conform with the Subcom- 
mittee recommendations, we chose the 
more fundamental diffuse source pro- 
cedure. Our equipment was specified 
to be null reading, in order to remove all 
questions of photocell response linearity 
and as nearly as possible to compare un- 
known with standard aperture simulta- 
neously, in order to avoid any possibility 
of faulty mechanical or electrical mem- 
ory. 

Both objectives have been realized 
(Fig. 1) by providing two apertures into 
the integrating box which illuminates 
the detector photocell. These aper- 
tures are alternately opened and closed 
thirteen times per second, so that the 
same total area is free to the lamphouse 
at all times first, all of one aperture, 
and as this closes, the other opens syn- 
chronously to completion. Thus when 
the flux incident on the two apertures is 
the same, there is a constant light level 
within the box. When, however, one 
aperture is blocked completely, the' light 
level varies as the size of the uncovered 
aperture, sinusoidally in this equipment. 



Allen E. Murray: T-Numbers Equipment 



81 



The two functions of the phototube 
and the electronic circuit are then to 
detect the state of light balance and to 
measure the degree of unbalance. Fig- 
ure 2 illustrates the principle of measure- 
ment, while Fig. 3 shows the schematic 
electric circuit. 

It is clear that balance comes about 
electrically when the amplitudes of the 
13-cycle signals arising from the two 
apertures are equal, for then, since they 



LAMPHOUSE 



OPAL GLASS 




STANDARD 
APERTURE 



A B 

INTEGRATING BOX 



PHOTOTUBE- 



TO AMPLIFIER 
Fig. 1. Schematic optical layout. 



are phased 180 apart, the resultant sig- 
nal is constant. 

The dominant aperture in the general 
case will phase the light signal. To de- 
termine which aperture this is and to 
produce a deflection at light balance, an 
auxiliary bipolar generator is synchro- 
nized with one aperture, and the measur- 
ing circuit designed to evaluate the sum 
of the light and generator signals. By 
adjusting the circuit properly, the indi- 
cator can be placed at midscale with 
only the alternating component of the 
generator effective so that deflections to 




TWO SIGNALS 
ADD IN PHASE 



LIGHT 
PATH A 



LIGHT 
PATH B 



GENERATOR 
SIGNAL IN 
PHASE WITH 
'A' AT ALL 
TIMES. SET 
FOR HALF SCALE 
DEFLECTION 



Fig. 2. Principle of measurement. 

When A > B resultant mixed signal 

is > half scale 

When A = B resultant mixed signal 

is half scale 

When A < B resultant mixed signal 

is < half scale 



RECEPTOR 
MULTIPLIER 




PR E AMR 
STAGE 




I3~TUNED 
AMPLIFIER 




I3^TUNED 
AMPLIFIER 




MIXER 1 (MEASURING 
STAGE [" H CIRCUIT 






V fi 






v in* 




040 







0~ SYN. 
GENERATOR 




I3~TUNED 
AMPLIFIER 









82 



Fig. 3. Schematic electrical circuit. 
January 1951 Journal of the SMPTE Vol. 56 




Fig. 4. General view of equipment. 




Fig. 5. Lens Mount and Standard Aperture Turret. 
Allen E. Murray: T-Numbers Equipment 



83 



the right mean that one aperture domi- 
nates, and to the left, the other. 

Moreover, by the use of the auxiliary 
calibrating apertures it is possible to 
calibrate the meter in terms of T-num- 
bers between each two consecutive full 
stops, and thus interpolate between 
stops for measurement or calibration 
purposes. This eliminates the need for 
neutral density filters. 

A general view of the equipment is 
contained in Fig. 4, and Fig. 5 shows the 
front of the integrating box. The lens 
standard carries a scale and vernier, 
which make it useful for measurement at 
finite conjugates. This scale is indis- 
pensable for maintaining or checking on 
the calibrations and sensitivity. 

The T-stops are defined by apertures 
placed a fixed distance from the inte- 
grating box aperture and carried on a 
turret plate. These are duplicated in a 
loose set of apertures fitting into an 
adapter in the lens standard. 

The illumination at present is pro- 
vided by a large lamphouse containing 
three 500-w projection lamps. It is 
coated white inside, and the front face 
is a large sheet of flashed opal glass. 
The uniformity of luminance of the face 
just meets the specifications contained 
in the Subcommittee Report. Some 
other arrangement doubtless would be 
safer in future equipments. 

This particular calibration unit has 
proved to be quite handy in practice, 
more than adequately sensitive with the 
focal plane apertures for the 35-mm and 
8-mm frames, with the 1P22 photomul- 
tiplier tube and three accelerating po- 
tentials, and self-contained in that it 
calibrates itself with the help of the aux- 
iliary apertures and basic instrumental 
dimensions. 

The reproducibility of measurement 
at all stops is of the order of less than 
1%, and the accuracy certainly well 
within the allowed =*=?% in illuminance. 

Acknowledgments: An enterprise such as 
this is the result of group effort, and there- 
fore it is necessary to distribute the credit 



for the design of this equipment. The 
mechanical design was ably carried out by 
R. Filsinger under the direction of O. 
Boughton, while the electronic circuit is 
the result of the joint efforts of K. H. 
Bloss and A. A. Shurkus, assisted by W. 
Ehlers. The mechanical assembly was 
under the direction of W. Guenther. 
The author also wishes to acknowledge the 
help given in conversations with his col- 
leagues, in particular Dr. K. Pestrecov and 
G. C. Wooters. 

Bibliography 

1. "Report of the SMPTE Subcommittee 
on Lens Calibration," Jour. SMPE, 
vol. 53, pp. 368-378, Oct. 1949. 

2. E. W. Silvertooth, "Stop calibration of 
photographic objectives," Jour. 
SMPE, vol. 39, pp. 119-122, Aug. 
1942. 

3. C. R. Daily, "A lens calibrating sys- 
tem," Jour. SMPE, vol. 46, pp. 
343-356, May 1946. 

4. M. G. Townsley, "An instrument for 
photometric calibration of lens iris 
scales," Jour. SMPE, vol. 49, pp. 
111-122, Aug. 1947. 

5. A. Odencrants, "A new method for the 
determination of the true transpar- 
ency of photographic objectives," 
Science et Industries Photograph- 
iques, vol. 6M, p. 57, Nov. 1926. 

6. J. Hrdlicka, "Measuring the effective 
illumination of photographic ob- 
jects," Jour. SMPE, vol. 14, pp. 
531-533, May 1930. 

7. J. H. Lambert, "Ostwald's Klassiker 

der Exakten Wissenschaften," Nr. 
31, Part 2, Chapters III and IV, 
Engelmann, Leipzig, 1892. 

8. D. B. McRae, "The measurement of 
transmission and contrast in opti- 
cal instruments," J. Opt. Soc. 
Amer., vol. 33, pp. 229-243, April 
1943. 

9. G. W. Moffitt, "Determining photo- 
graphic absorption of lenses," J. 
Opt. Soc. Amer., vol. 4, pp. 83-90, 
May 1920. 

10. D. B. Clarke and G. Laube, "Twenti- 
eth Century camera and accesso- 
ries," Jour. SMPE, vol. 36, pp. 50- 
64, Jan. 1941; also U. S. Pat. 
2,334,906, 1943. 



January 1951 Journal of the SMPTE Vol. 56 



11. L. T. Sachtleben, "Method of cali- 
brating lenses," U. S. Pat. 2,419,- 
421, 1947. 

12. I. C. Gardner, "Compensation of the 
aperture ratio markings of a photo- 
graphic lens for absorption, reflec- 
tion, and vignetting losses," Jour. 
SMPE, vol 49, pp. 96-110, Aug. 
1947: also /. Res. Nat. Bur. Stand., 
vol. 38, pp. 643-650, June 1947. 

13. F. G. Back, "A simplified method for 
precision calibration of effective f- 
stops," Jour. SMPE, vol. 49, pp. 
122-130, Aug. 1947. 

14. E. Berlant, "A system of lens stop 
calibration by transmission," Jour. 
SMPE, vol. 46, pp. 17-25, Jan. 
1946. 

15. A. E. Murray, "The photometric 
calibration of lens apertures," Jour. 
SMPE, vol. 47, pp. 142-151, Aug. 
1946. 

The following three references are of 
historical interest : 

16. C. Forch and E. Lehmann, "Die Licht- 
verluste in photographischen Ob- 
jectiven," Kinotechnik, vol. 10, pp. 
3-7, Jan. 5, 1928. 

17. A. Klughardt and H. Otto, "Licht- 

starkemessung hinter photograph- 
ischen Objectiven den Kleinbild 
Kamera," Photographische Indus- 
trie, vol. 34, pp. 608-610, May 
1936. 

18. A. Kochs, "Lichtstarke und relative 
Offnung von Kine-Objectiven," 
Kinotechnik, vol. 22, pp. 154157, 
Nov. 1940. 



Discussion 

M. C. TOWNSLEY: Is your instrument 
primarily intended for calibrating aper- 
tures or measuring apertures? 

MR. MURRAY: Actually, of course, the 
equipment does both. It was designed 
primarily to calibrate, but we picked up fa- 
cilities here and there in the course of the 
design of the equipment. We are more 



than pleased that it will serve both func- 
tions. 

MR. TOWNSLEY : It looked from the way 
it was laid out that it could do both and 
probably do them quite well. 

MR. MURRAY: We have felt from the 
very beginning that calibration alone 
would not be sufficient. We wanted to be 
able from our own equipment, independ- 
ently of any other, or from fundamental 
geometry and mechanical construction, to 
determine that the calibration is done 
properly. 

MR. TOWNSLEY: In calibration, I am 
thinking of starting with an unknown lens 
and marking a set of apertures in T-stops. 

MR. MURRAY: We can do that very 
well for the standard apertures on the tur- 
ret plate. Their distance from the obscur- 
ing aperture in the lamphouse is known to 
around two tenths of one per cent. The 
diameter of each aperture has been meas- 
ured along at least four meridians, so that 
we know they are circular. Their edges 
have been specially treated to cut down re- 
flection. We put in what refinements we 
could see. 

MR. TOWNSLEY: I rather hope for the 
future of the T-stop system that you will 
do a great deal of original calibration on 
customer lenses on that system actually 
mark them in T-stops. 

MR. MURRAY: I am authorized to say 
that this equipment will be used also for 
customer lens calibration. Our sales de- 
partment has just let me know that we 
are ready to undertake this sort of work. 
You personally might bfc interested to 
know that we have had the opportunity to 
look over some of your lenses, and we are 
very pleased to note that we agree very 
closely. 

This paper was prompted by some ques- 
tion as to whether one method is better 
than another. We must say "no." 
There is no visible justification for setting 
up a standard around one particular de- 
vice or method. Any equipment is satis- 
factory as long as it conforms to physical 
principles and the requirements of sound 
engineering. 



Allen E. Murray: T-N umbers Equipment 



85 



The Differential Carbon-Feed System 
for Projection Arc Lamps 

By Arthur J. Hatch 



There is a growing recognition of the fact that to obtain constant screen 
color and light intensity, the position of the positive carbon must be 
maintained automatically in relation to the projection lamphouse optical 
system. In the development and application of such a control feature, the 
requirements of carbon-feed systems have been reviewed. The differential 
carbon-feed system seems to meet these requirements, and considerations 
pertaining to the application of the differential feed system with automatic 
positioning to an angular trim burner will be related. 



rriHE CHALLENGE offered by present 
_l large screens, and the demand for 
higher picture brilliancies have led to 
the wide adoption of high-speed projec- 
tion lamphouse optics, and carbons 
with higher intrinsic brilliancy. 

With these the allowable tolerance in 
carbon crater position has been reduced 
by the use of the higher-speed lamp- 
house optics, while the difficulty of main- 
taining the arc crater at a given position 
has been increased by the high bright- 
ness carbons with their higher burning 
rates. These higher burning rates are 
unfortunately accompanied by greater 
fluctuations of burning rate with small 
current changes. These factors have 
made it desirable to incorporate auto- 
matic means in the carbon feed to main- 
Presented on April 27, 1950, at the Soci- 
ety's Convention at Chicago, 111., by 
Arthur J. Hatch, The Strong Electric 
Corp., 87 City Park Ave., Toledo 2, Ohio. 



tain the position of the positive crater 
accurately to the lamphouse optical 
system. 

This problem of providing automatic 
positioning to the positive crater "of 
high-intensity projection arc lamps has 
necessitated a review of the require- 
ments for carbon feeds, as such a posi- 
tioning control cannot be conveniently 
or effectively inserted into the type of 
feed mechanisms in general use at pres- 
ent. Accordingly, to utilize an auto- 
matic positioning device it has been 
necessary to develop a new carbon-feed 
system. 

To anyone not especially acquainted 
with operation or design of projection 
arc lamps, the feeding of the carbons 
would seem a very simple matter that 
could readily be solved by merely ar- 
ranging a motor drive to both carbons. 
However, as it is with so many other 
seemingly simple problems, this subject 
is not altogether simple when the com- 
plete requirements are known. 



January 1951 Journal of the SMPTE Vol. 56 



Requirements of Carbon-Feed 
System 

We find that the principal end results 
desired are uniform and constant inten- 
sity of screen illumination with constant 
color temperature. These results should 
be obtained through a carbon-feed sys- 
tem that has simple control adjust- 
ments and which is capable of self-com- 
pensation for changes in the variables, 
without attention from the projection- 
ist. 

Upon examining these requirements 
for a feed system, we find that the major 
electrical controlling factor necessary to 
obtain constant screen illumination, 
with a given carbon trim, is constant arc 
amperage. 1 With proper arc circuit bal- 
last, the arc amperage will assume a 
value such that the sum of the positive 
and negative carbon-burning rates, at 
that arc current, equals the sum of the 
positive and negative feed rates. Then 
assuming for the moment that the car- 
bon-burning rates are constant for a 
given current, it will be readily seen 
that a constant total feed rate will pro- 
vide most even illumination. 

Therefore, a very simple carbon-feed 
mechanism could be constructed which 
would advance the relative positions of 
the carbon holders one to the other at 
the constant rate necessary to maintain 
the desired current. 

The negative carbon could stand still 
and the positive carbon could be ad- 
vanced at a rate equal to the total burn- 
ing rate of both carbons ; or the positive 
could stand still and the negative could 
advance at the total rate. Any number 
of positive and negative feed ratios 
could be used as long as the combined 
feed added to the figure desired for total 
feed. 

This simple feed, however, would not 
take into account the fact that to utilize 
the illumination from the carbon arc 
for projection, the positive crater must 
be kept at the exact entrance focal posi- 
tion of the lamphouse optical system. 
It is, therefore, necessary to make provi- 



sion to divide the total feed into posi- 
tive and negative feeds, in a proportion 
exactly equal to the positive and nega- 
tive burning rates at the particular cur- 
rent desired, in order to maintain the 
position of the positive crater to the op- 
tical system. 

This division of the total feed into its 
components needs to be flexible, unless 
the lamp is to be burned at a single cur- 
rent, as the ratio between positive and 
negative burning rates varies consider- 
ably through the current range of the 
carbons. 2 

The operation of this ratio-fixing con- 
trol should not affect the sum total feed 
rate of the positive and negative car- 
bons. For this reason a ratio-changing 
system is necessary in which, if the 
negative feed is slowed down, the posi- 
tive feed is increased simultaneously so 
that total carbon feed and constant 
current are maintained. 

An Ideal Feed System 

From the foregoing it is easy to draw 
a conclusion that an ideal feed system 
would be one in which one control deter- 
mined the total feed and the other con- 
trol determined the ratio between posi- 
tive and negative feeds. With a system 
of this type, the total feed control could 
be set for the desired amperage and the 
ratio control adjusted until the feed 
ratio matched the burning ratio. This 
second adjustment would not affect the 
feed-control setting. 

Thus, for example, with a 7-mm nega- 
tive and 8-mm positive copper-coated 
high-intensity trim the total burning 
rate for both carbons at 70 amp is ap- 
proximately 20 in./hr. The current 
selector would be set to produce this 
total rate of feed. Then the ratio con- 
trol would be adjusted, until the posi- 
tion of the burning tip of the positive 
carbon in relation to the optical system 
was correct and its relative movement 
reduced to zero. It thus might be found 
necessary to adjust the ratio control 



Arthur J. Hatch: Differential Carbon Feed 



87 



setting so that the negative feeds 4 in./ 
hr, and the positive, 16 in./hr, or the 
negative might be fed 4J^ in ,/hr, and the 
positive, 15% in./hr. In either case 
the total feed would remain at 20 in./hr. 
and the arc current at 70 amp. 

This ideal feed system is analogous 
to that of a mechanical differential sys- 
tem, a common type of which is found 
in the rear drive of automobiles. Here 




Fig. 1. Carbon position-detecting op- 
tical system showing prism lens and 
bimetallic switch. 



the speed of the torque tube drive shaft 
is analogous to the total feed of both 
carbons. The sum of the drive of the 
two rear axles is a constant for constant- 
torque tube drive and the ratio between 
axles can be varied by restraining one 
wheel in which case the other wheel 
turns faster. 

A practical embodiment of this ideal 
feeding system can be realized with the 
use of a two-motor drive. One motor, 
which is the feed motor, drives both 
carbons through a differential gear drive. 
The second, or rate-control motor, is 
connected preferably in the negative 
drive. The resultant difference in drive 
between the feed motor and the rate- 
control motor is transmitted to the 
positive carbon feed. Gear ratios are 
chosen so that the resultant total feed 
of both carbons is, at all ratios, a con- 
stant as determined by the speed of the 
main drive. 




Fig. 2. General view of differential feed burner from operating side 
showing positive and negative feeds and the single adjustment control. 



January 1951 Journal of the SMPTE Vol. 56 



Need for Automatic Positioning 

This feeding system and almost all 
present arc feeding systems make an 
assumption that there will be little or no 
variation in arc gap length, carbon- 
burning rate or power supply voltage. 
However, in practical experience these 
ideal conditions are seldom satisfied. 

Variations in carbon-burning rates 
and ratios at a given current, of course, 
directly reflect a change of position of 
the arc with respect to the lamphouse 
optical system. Arc-gap lengths at 
identical currents and even with con- 
stant applied arc voltage will vary from 
trim to trim and even within a trim. 
With constant arc current, the depend- 
ent variable that compensates for 
variation in arc supply voltage is the 
arc-gap length. As the positive carbon 
has the highest burning rate (being ap- 
proximately 2 to 8 times that of the 
negative carbon), the major adjustment 
in position for variations in arc-gap 
length occurs in the position of the posi- 
tive carbon. Thus, variations of arc 
voltage or gap length directly affect the 
position of the positive crater in relation 
to the optical system. 

Therefore, to adopt the ideal carbon- 



feed system to these practical considera- 
tions, there must be introduced an ele- 
ment that will maintain the positive 
crater at the optical focal point regard- 
less of variation in arc gap or burning 
rate. 

It is, therefore, practical to introduce 
a carbon crater position-detecting and 
ratio control-actuating mechanism into 
this system to accomplish this end. The 



0< A*M 

ifOjU RELAY 




- O 



Fig. 3. Simplified arc control 
circuit diagram. 



1 




Fig. 4. General view of burner from nonoperating side showing 
motors and bimetallic switch behind left motor. 



Arthur J. Hatch: Differential Carbon Feed 



89 



bimetallic element with its* ruggedness 
and simplicity seems to be most practi- 
cal for this position detector. 3 

This bimetal switch is simply arranged 
to shunt out a series resistance in the 
ratio-motor field circuit. With all re- 
sistance shunted out, the ratio motor 
runs at a speed such that the negative 
carbon is fed at a rate below its burning 
rate, and the positive is fed at a rate 
above its burning rate. When the re- 
sistance is inserted by action of the bi- 
metal switch, the negative is fed at a 
rate above, and the positive at a rate 
below its burning rate. 

Total rate of feed at any selected 
amperage is obtained from the main- 
drive motor, and the position of the posi- 
tive carbon is accurately maintained 
with the controlled variation of the 
ratio motor. 

Angle-Trim Lamp-Feed 
Considerations 

With the use of angle-trim lamps, the 
general considerations for constant il- 
lumination remain the same with the ex- 
ception that to maintain this even il- 
lumination, the feed rate of the negative 
has to be corrected for its angular direc- 
tion before it can be added to the posi- 
tive to obtain the value for combined 
total feed. 

It has been confirmed by experiment 
that, within a reasonable limit of move- 
ment, if the positive carbon is underfed 
a certain amount, X, an overfeed of the 
negative equal in amount to X cosine 
a will maintain constant arc current, 
where a is the depression angle of the 
negative in relation to the positive. 

Taking advantage of the uniform and 
predictable speed characteristics of the 
d-c shunt motor, it is possible to design 
an electrical differential motor feed cir- 
cuit whereby the use of the mechanical 
differential is eliminated. With this 
arrangement, each carbon is driven by a 
separate motor. Such a system, with- 
out an automatic position-control 
switch, would contain two controls, each 



consisting of two rheostats connected in 
mechanical tandem. Each of the rheo- 
stats in the total feed-rate control would 
be connected in the field circuit of its 
respective motor, and the resistance 
values arranged so that the carbon-feed 
speeds were changed approximately in 
their correct values throughout the 
entire current range of the carbons. 

The ratio-control rheostats would be 
connected in the two-motor field circuits 
in such a manner that as the ratio con- 
trol was advanced, the positive feed 
motor would be slowed and the negative 
feed motor would be speeded the correct 
amount to maintain the same current 
in the lamphouse. 

For automatic positioning, the bi- 
metallic element would be arranged to 
shunt in and out portions of this ratio- 
control rheostat. The general optical 
arrangement for projecting the energy 
image of the positive carbon and flame 
to the bimetallic switch is shown in Fig. 
1. The 90 prism with a lens ground in 
one face is used to direct the side view 
of the arc to the glass-enclosed bimetallic 
switch. 

Single-Feed Control 

It is possible to obtain d-c shunt 
motors with speed characteristics such 
that as the arc voltage is raised, con- 
sistent with higher arc currents, the 
negative feed motor will increase in 
speed approximately the right amount to 
compensate for the increased negative 
burning rate. 

This fact, in conjunction with the use 
of a fairly large speed differential on 
both motors, controlled by means of the 
position-sensitive device, has enabled 
considerable simplification of the con- 
trol circuit. 

The net result has been the develop- 
ment of a circuit in which complete 
control of both carbon feeds throughout 
their entire amperage range has been 
accomplished with but a single lamp- 
house feed-control adjustment. This 
control is in the form of a single rheo- 



90 



January 1951 Journal of the SMPTE Vol. 56 




Fig. 5. (a) The arc burning with no 
air supplied from jet showing the 
characteristic long- tail flame reaching 
toward the optical system. 



Fig. 5. (b) The burning arc showing 
how the application of air from the 
jet shortens and redirects the flame. 



stat which is provided with a pointer 
and a scale indicating arc amperages. 
The general arrangement of components 
of a burner incorporating this two- 
motor, single-control feed system as 
viewed from the operating side is 
shown in Fig. 2. A simplified wiring 
diagram of this system is shown in Fig. 
3. 

The rheostat is connected in the posi- 
tive feed motor field circuit and has a 
value sufficient to control the feed of the 
positive carbon through a range of from 
14 to 32 in./hr. 

The bimetallic switch is connected in 
such a manner that in its open position, 
a resistor is inserted in the positive field, 
and a resistance is shunted out in the 
negative field, thus speeding the positive 
and simultaneously slowing the nega- 



tive. When the bimeta'lic switch is 
closed by reason of the positive carbon 
position being slightly too near the opti- 
cal system, the resistor in the positive 
field circuit is shunted, and the resistor 
is simultaneously inserted in the nega- 
tive field circuit, thus slowing the posi- 
tive and speeding the negative. 

The positive motor will change speed 
sufficiently with this cycling to change 
the feed rate by approximately 4 in./hr 
from fast to slow rate. With the nega- 
tive carbon being depressed at an angle 
of 52, its feed rate is arranged to change 
4 X cos 52, or approximately 2.5 in./ 
hr from fast to slow. 

When the arc current selector rheo- 
stat is set at the desired current, the 
positive motor assumes a speed, such 
that the average speed between high- 



Arthur J. Hatch: Differential Carbon Feed 



91 



and low-cycle speeds is equal to the 
average burning rate of the positive car- 
bon at the selected current. 

If the arc current at a particular in- 
stant is slightly less than the selected 
current, the positive burning rate will be 
slightly lower than the average positive 
feed rate. Consequently, the arc posi- 
tion-control switch will remain in the 
low-speed positive feed position longer 
at a time, than in the high-speed posi- 
tive feed position. This will cause the 
negative to be fed at a greater average 
rate than it is being consumed, thereby 
shortening the arc gap, and raising the 
current, until an equilibrium condition 
is reached, at which the average nega- 
tive and positive burning rates equal 
the average feed rates. This will be 
realized at approximately a 50% division 
of time on high and low speeds. 

If the arc current, and consequently 
the positive burning rate is higher than 
the selected rate, the arc position-control 
switch will remain in the high-speed 
position longer at a time than in the low- 
speed position. This will cause the 
negative to be fed at a lower than aver- 
age rate, thereby lengthening the arc 
gap until equilibrium is reached. 

Slow changes in power supply voltage 
are compensated for by the automatic 
resulting change in arc-gap length, but 
with the continual maintenance of the 
positive crater at the required position. 



Miscellaneous Features 

Secondary considerations in connec- 
tion with the realization of the two- 
motor automatic positioning drive in- 
clude the provision of centrifugal fans 
on each of the motors (see Fig. 4). These 
fans exhaust into the burner base en- 
closure from where the air is directed 
up through the rotating positive feed 
head, and against the negative feed 



head, thereby keeping these parts at low 
operating temperatures. 

Immediately above and parallel to the 
negative carbon is located a jet tube 
which directs a stream of air at the arc 
tail flame immediately above the crater. 

This device has several useful func- 
tions in that it shortens and redirects the 
tail flame away from the reflector, as 
shown in Fig. 5. The white ash product 
of combustion of the arc is blown away 
from the reflector thereby eliminating 
deposit on the reflector and the conse- 
quent breakage caused by heat differen- 
tials. 

Another benefit derived from the air 
jet is that it supplies enough additional 
air to the vicinity of the arc that upon 
striking the arc, the soot particles 
are consumed instead of being released 
to the reflector surface, or lamp-house 
interior. 

Finally, the air jet causes the blend- 
ing of the negative and positive flames 
and results in excellent stabilization of 
the arc without the use of an auxiliary 
magnetic field. Thus, with the embodi- 
ment of the differential concept of car- 
bon feed which was developed for the 
purpose of obtaining uniform feed in 
conjunction with automatic positioning 
of the positive crater, it is possible to 
stabilize the burning of the arc and keep 
the products of combustion from the 
lamphouse optical system. 

References 

1. D. B. Joy and E. R. Geib, "The non- 
rotating high-intensity d-c arc for pro- 
jection," Jour. SMPE, vol. 24, pp. 
47-62, Jan. 1935. 

2. "Report of the Projection Practice 
Committee," Jour. SMPE, vol. 24, pp. 
35-46, Jan. 1935. 

3. D. J. Zaffarano, W. W. Lozier and D. B. 
Joy, "Improved methods of controlling 
carbon arc position," Jour. SMPE, 
vol. 37, pp. 485-501, Nov. 1941. 



92 



January 1951 Journal of the SMPTE Vol. 56 



Bibliography 

on High-Speed Photography 

Including Schlieren and Cathode-Ray Oscillograph Photography 



Contents I. General 94 

II. Cameras 95 

III. Lighting 97 

IV. Oscillography 98 

V. Schlieren 101 

VI. Technical and Techniques. . 102 

VII. X-Ray Ill 



This bibliography was compiled by Miss Elsie Garvin, Librarian, Re- 
search Library, Eastman Kodak Co., Rochester, N.Y., and was recom- 
mended for publication in the Journal by the Society's High-Speed Pho- 
tography Committee. Those who reviewed the 600 items suggested that 
they be arranged by subject and that certain entries be expanded by anno- 
tation. John H. Waddell, Chairman of the Committee, undertook the job 
of classification; many more items were added, dating up to July, 1950; 
and manuscript was released on December 7, 1950, for publication. 

The Society has previously published two reprint volumes on high-speed 
photography and this bibliography will be the last item in Vol. 3 which 
will include also those papers on the subject which appeared in the Journal 
during 1950. 

To expand its usefulness the third volume will include a cumulative table 
of contents showing the titles and authors of all papers in Vols. 1 through 
3. It will cover all articles on the subject which appeared in the Journal 
beginning with the special issue, Part II of the March, 1949, Journal. 



January 1951 Journal of the SMPTE Vol. 56 93 



I. GENERAL 



L' Ultra-cinema, P. Nogues, Cinemat. franc., 
9: 29-32, Apr. 30, 1927. 

Der Zeitdehner, sein Bau und seine Anwend- 
ung, R. Thun, Kinotechnik, 10: 119-26, 
Mar. 5, 1928. (General discussion of in- 
struments working between limits of 100 
1,000,000 per sec.) 

Rapidkameras, Lichtbildbiihne, 21: 16-19, 
Nov. 3, 1928. (Historical.) 

Kinematographie auf ruhendem Film und 
mit extrem hoher Bildfrequenz, C. Cranz 
and H. Schardin, Zeit. Physik, 66: 147-83, 
#3 /4, 1929. (1 flO to 1/3 millionth of a 
sec.) 

Theorie der Thunschen Zeitdehners und ihre 
Anwendung in der Augnahmepraxis, W. 
Ende, Zeit. tech. Physik, 11: 394-^02, #10, 
1930. 

Premiere Essais de Cinematographic UUra- 
rapide, A. Magnan, Hermann et Cie, 
Paris, 1932. (History.) 

Kinematographische und photographische 
Untersuchungsmethoden schnell verlau- 
fender Vergange, F. Beck. Ber. Int. Kong. 
Phot. Dresden, 1931, 261-65, 1932. 

High-Speed Motion Pictures, H. E. Edgerton, 
Elec. Eng., 64: 149-53, Feb. 1935. 

Speed Photography, New Photo-Miniature 
(old series #206), New series #1, 1-32, 
1935. 

High-Speed Photography, H. E. Edgerton, 
K. J. Germeshausen and H. E. Grier, 
Phot. J., 76: 198-204, Apr. 1936. 

Sound Waves Their Shape and Sound, D. C- 
Miller, MacmiUan, New York, 1937- 
(Contains a chapter on Electric-Spark Photo- 
graphy of Sound Waves.) 

Entwicklung der Hochfrequenzkinemato- 
graphie, F. E. von Eckard, Filmtechnik, 13: 
121-23, July 6, 1937. 

High-Speed Photographic Methods of Meas- 
urement, H. E. Edgerton, K. J. Germes- 
hausen and H. E. Grier, /. Appl. Phys., 8: 
2-9, Jan. 1937. 

Flash! Seeing the Unseen by Ultra High-speed 
Photography, H. E. Edgerton and J. R. 
Killian, Jr., Hale, Cushman and Flint, 
Boston, 1939. (Bibliography.) 

Tampering with Time, Gjon Mili, Photo 
Technique, 1: 4-9, Aug. 1939. 

Action Photography, G. Denes, Fountain 
Press, London, 1940. 

Photographische Messtechnik, L. Fink, R. 
Oldenbourg, Munchen, 1940. 

Photographing Action, V. De Palma, Ziff- 

Davis, Chicago 1940. 

High-Speed Photography, C. Tuttle and F. 
Brown, Complete Photog., 6: 1965-68, 
Issue 30, 1942; 6: 1969-74, Issue 31. 
1942. 



High-Speed Flash Photography, H. L. 

Westrate, Intern. Phot., 14: 4-5, Sept. 

1942. 
Photographic Analysis of Motion, E. R. 

Davies, Nature, 162: 261-64, Sept. 4, 

1943. 
High-Speed Photography. A Monograph, L. 

Wells, Australasian Photo-Rev., 61: 317-23. 

25, Oct. 1944. 
Photography in Engineering, C. H. Tupholme, 

Faber and Hyperion Ltd., London, 1945. 

(Chapter called "High-Speed Photography," 

pp. 99-131.) 
The Scientist in Wartime, E. Appleton, 

Engineer, 180: 432-33, Nov. 30, 1945. 
Speed Camera. The Amateur Photography of 

Motor Racing, E. S. Tompkins, G. T. 

Foulis, London, 1946. 
High-Speed Motion Picture Photography, 

H. M. Lester, /. Biol. Phot. Assoc., 14: 

107-18, Mar. 1946. 
High-Speed Photography, (Alpha Photo 

Lab), The Cinema: 45, Apr. 3, 1946. 
Photographic Aspects of High-Speed Re- 
cording, W. F. Berg, Phot. J., 86B: 154-58, 

Nov. 1946. 
Flash in Modern Photography, William 

Mortensen and D. M. Paul, Camera Craft, 

San Francisco, 1947. 
Flash Photography, Gordon Parks, Franklin 

Watts, New York, 1947. 
Synchronized Flashlight Photography, G. L. 

Wakefield and N. W. Smith, Fountain 

Press, London, 2nd ed., 1947. 
Symposium on High-Speed Photography, 

Brit. J. Phot., 94: 44, Jan. 24, 1947. 
The Past, Present and Future of High-Speed 

Photography, H. E. Edgerton, PSA Jour., 

13: 437-38, July 1947. 
Graphic Graflex Photography, Willard D. 

Morgan and H. M. Lester, Morgan and 

Lester, New York, 8th ed., 1948. 
Le cinema scientifique francaise, P. Theven- 

ard and G. Rassel, La Jeune Parque, Paris, 

1948. (Etude analytique des mouvements 

rapides, pp. 3-39; historical, high-speed and 

very high-speed.) 
"Strobe" the Lively Light, H. Luray, Camera 

Craft, San Francisco, 1949. 
What is High-Speed Photography, M. L. 

Sandell, Jour. SMPE, 52: 5-7, Mar. Pt. 2, 

1949. 
A Photographic Study of Rapid Events, W. D. 

Chesterman. (Book in preparation for 

Oxford University Press.) 
Photographic Instantanee et Cinematographic 

Ultra-Rapide, P. Fayolle and P. Naslin, 

Editions de la Revue d'Optique, Paris, 1950. 
A Survey of High-Speed Motion Picture 

Photography, K. Shaftan, Jour. SMPTE, 

64: 603-26, May 1950. 



94 



January 1951 J ournal of the SMPTE Vol.56 



II. CAMERAS 



1,000,000 Pictures per Minute, C. F. Jenkins, 
Trans. SMPE, No. 13: 69-73, 1921. 

Analysis of Motion, C. P. Watson, Trans. 
SMPE, No. IS: 65-58, 1921. (Novograph 
high-speed motion picture camera. 125-160 
pictures per sec.) 

Motion Picture Camera Taking 3,200 Pic- 
tures per Second, C. F. Jenkins, Trans. 
SMPE, No. 17: 77-78, 1923. 

The Heape and Grylls Camera, C. N. Ben- 
nett, Kinemat. Weekly, 87: 76, May 15, 
1924. 

A Camera for Studying Projectiles in Flight, 
H. L. Curtis, W. H. Wadleigh and A. H. 
Sellman, Tech. Papers Bur. Stand., #255, 
18: 189-202, 1924. 

Die Lyta-Spiegelreflex-Kamera mit Hoch- 
frequenzeinrichtung, O. C. Tauern, Kino- 
technisches Jahrbuch, 55-59, 1925-26. 

The Jenkins Chronoteine Camera for High- 
Speed Motion Studies, C. Jenkins, Trans. 
SMPE., 10: 25-30, #25, May 1926. 

The Heape and Grylls' Machine for High- 
Speed Photography, W. H. Connell, J. 
Sci. Instr., 4: 82-87, Dec. 1926. 

Askania-Hochfrequenz-Modell 27, H. Friess 
Filmtechnik, 3: 62-64, Feb. 19, 1927. 

Der Thun'sche Zeitdehner. Model NSOKT 
3, A. v. Barsy, Kinotechnik, 9: 348-52, 
July 5, 1927. 

Appareil a grande vitesse Gaumont pour 
prise de vues cin6matographiques, Etab- 
lisements Gaumont, Bull. soc. frang. phot., 
(3) 14: 204-7, July 1927. 

A Novel High-Speed Camera, E. B. Wed- 
more, J. Sci. Instr., 4.- 345-47, Aug. 1927- 
(A variation of Heape and Grylls' high-speed 
camera.) 

Ein Report-Apparat, Filmtechnik 3: 314-15, 
Aug. 20, 1927. (Gaumont camera, 160 ex- 
posures per sec., works forward or back- 
ward.) 

Fin Rapidkamera vom Jahre 1897, A. Las- 
sally, Kinotechnik, 9: 588-95, Nov. 20, 

1927. (100 pictures per sec.) 

Die Anwendung des Zeitdehners beim Stu- 
dium von Kavitations-erscheinungen, H. 
Mueller, Kinotechnik, 10: 462-65, Sept. 5, 

1928. (75,000 exposures per sec.) 

A New Ultra-Speed Kinematographic 
Camera Taking 40,000 Photographs per 
Second, T. Suhara, Proc. Imperial Acad. 
Japan, 5: 334, Oct. 1929. 

Der Thun'sche Zeitdehner des Institutes 
fur Kleinzeitforschung, Kinotechnik, 11: 
145-47, Mar. 20, 1929. (Thun's camera, 8 
to 10 meters of film per sec. or 4000 pictures 
per sec. with exposures as short as 1 175,000 
per sec.) 



The Fearless High-Speed Camera Movement, 
Intern. Phot., 1:41, Oct. 1929. 

Die Neue Zeitlupe, H. Joachim, Filmtechnik, 
6: 10, Jan. 11, 1930. (Zeiss Ikon improve- 
ment of old Ernemann camera.) 

Photographing Sound, C. Neblette, Photo 
Era, 64: 160-63, #3, 1930. (Foley's ap- 
paratus for spark photography is illus- 
trated.) 

New Ultra-Speed Kinematographic Camera 
Taking 40,000 Photographs per Second, T. 
Suhara, N. Sato and S. Kamei, Report of 
the Tokyo University Aeronaut. Research 
Inst., #60, May 1930. (Translated by P. 
Schrott from above as: Eine neue Hochfre- 
quenz-Kino-Kamera fur 40,000 Bilder je 
Sekunde. Kinotechnik, 12: 495-97, Sept. 
20, 1930.) 

Electrooptical Shutter Its Theory and 
Technique, F. G. Dunnington, Phys. Rev., 
38: 1506-34, Oct. 15, 1931. 

High-Frequency Camera for the Study of 
Electrical Discharges, Photo Era, 67: 102, 
Aug. 1931. (Twenty-two matched lenses 
focused on an 8 X 11 in. plate or cut film 
holder. Finally arranged for stereoscopic 
effects.) 

Moreno-Snyder Camera Makes High Speed 
for Trick Shot, Amer. Cinemat., 12: 43, 
Sept . 1 93 1 . (1 440 feet per minute with non- 
intermittent camera.) 

Latest Newman-Sinclair Auto-Kine Models, 
Kinemat. Weekly, 178: 38-39, Dec. 17, 
1931 . (10 to 24 frames per sec.) 

Cinematographic jusqu'a 12,000 Vues per 
Second, A. Magnan, Hermann et Cie, 
Paris, 1932. 

Stroboscopic and Slow-Motion Pictures by 
Means of Intermittent Light, H. E. 
Edgerton, Jour. SMPE, 18: 356-64, Mar. 
1932. 

The Problem of Motion Picture Projection 

from Continuously Moving Film, F. E. 

Tuttle and C. D. Reid, Jour. SMPE, 20: 

3-30, Jan. 1933. 
Theodolite cinematographique a grande 

vitesse Debrie, Rev. d'Optique, 12: 222-25, 

May 1933. 

A Non-Intermittent High-Speed 16-mm 
Camera, F. E. Tuttle, Jour. SMPE, 21: 
474-77, Dec. 1933. 

Stroboscopic Light High-Speed Motion 
Pictures, H. E. Edgerton and K. J. Ger- 
meshausen, Jour. SMPE, 23: 284-98, Nov. 
1934; Trans. Amer. Inst. Chem. Eng., 30: 
420-37, 1934. 

Zeiss Ikon High-Frequency Camera, Brit. J. 
Phot., 82: 617-18, Sept. 27, 1935. 



Bibliography on High-Speed Photography 



95 



Uber den Verdampfungsvorgang nach kine- 
matographischen Aufnahmen an Dampf- 
blasen, W. Fritz and W. Ende, Physik, Zeit. 
37: 391-401, #11, 1936. (Modification of 
Thun's high-speed camera; process of evap- 
oration shown by cinematograph pictures of 
bubbles of steam.) 

Le cinema scientifique en France et ses 
realisations techniques, P. Michaut, Tech. 
cinemat., 8: 757-58, Oct. 1936. (J. Pain- 
leve's stop-motion picture camera for color 
cinematography.) 

100 Bilder/Sek. mit Achterfilm!, Film fur 
Alle, 10: 140-41, May 1936. (The new 
Cine-Nizo 8ZD 8-mm camera taking up to 
100 frames per sec.) 

Photographic "Judge" at the Olympic 
Games, Brit. J. Phot., 88: 688-89, Oct. 30, 
1936. (Zeiss-Ikon duplicate stereocinemato- 
graph apparatus.) 

Photographic Race-Timing Equipment, F. 
E. Tuttle and C. H. Green, Jour. SMPE, 
27: 529-36, Nov. 1936. 

Studying High-Speed Operations: New 
Camera to Take 2,000 Exposures a Second, 
Ind. Eng. Chem. News ed., 14: 474, Dec. 
10, 1936. (Motion picture camera with 
quartz lenses.) 

A Rotating Drum Camera for Photographing 
Transient Phenomena, R. Lambert, Rev. 
Sci. Instr., 8: 13-15, Jan. 1937. 

Drum Camera for Recording Transient 
Electrical Phenomena, F. E. Lutrin, J. 
Sci. Instr., 14: 209-12, June 1937. 

Six Thousand Images a Second Can Be 
Taken with This Camera, Photography, 6: 
47, Feb. 1937. (8-Mm motion picture 
camera developed in France by Merlin and 
Gerin buiU to take 6000 images per sec.) 

Der Zeitdehner der Technik fur 80,000 bis 16 
Aufnahmen in der Sekunde, Phot. Ind., 35: 
391-94, Mar. 31, 1937. (German A.E.G. 
camera 16 to 80,000 images per sec.) 

High-Speed Motion Picture Photography 
Applied to Design of Telephone Apparatus, 
W. Herriott, Jour. SMPE, SO: 30-37, Jan. 
1938. 

8-Mm. Ultra-Rapid Camera, Kinemat. 
Weekly, 262: 36, Dec. 15, 1938. (Merlin 
Gerin's camera.) 

Vinten High-Speed Camera, D. H. Geary, 
Phot. J., 79: 291-92, Apr. 1939. 

Der neue Askania Zeitraffer, A. Jotzoff, 
Kinotechnik, 22: 15-17, Feb. 1940; cf. 
Askania bringt neuen Universal Zeit- 
raffer, H. Linke, Filmtechnik, 16: 61-62, 
May /June 1940. 

A 120,000 Exposure-per-Second Camera, D. 
C. Prince and W. K. Rankin, Gen. Elec. 
Rev., 42: 391-93, Sept. 1939. 



Cameras for Lightning Studies, C. J. Kettler, 

Photo Technique, 2: 38-13, May 1940. 
A Concise Report on a High-Speed Camera 

for Simultaneous Photographic and Oscil- 

lographic Records, H. W. Baxter, J. Sci. 

Instr., 19: 183-84, Dec. 1942. 
Fastax: Ultra-High-Speed Motion Picture 

Camera, H. J. Smith, Bell Lab. Record, 22: 

1-4, Sept. 1943. 
Study of High Explosives by High-Speed 

Photography, R. W. Cairns, Ind. Eng. 

Chem., 36: 79-85, Jan. 1944. 
High-Speed Cinematograph Camera, P. S. 

H. Henry, J. Sci. Instr., 21: 135-41, Aug. 

1944. 

The Eastman High-Speed Camera, Type 
III, J. L. Boon, Jour. SMPE, 43: 321-26, 
Nov. 1944. 

High-Speed Cameras, E. D. Eyles, J. Brit. 
Kinemat. Soc., 7: 84-91, July-Sept. 1944; 
Amer. Cinemat., 25: 372-73, 380, 392, 
Nov. 1944. 

The Ribbon-Frame Camera (High-Speed 
Still Camera), F. Reck, Bell Lab. Record, 
23: 40-15, Feb. 1945. 

Sur un cinematographic rapide pour pellicule 
de 9 mm de large donnant de 1500 a 2000 
images par seconde, A. Magnan, Compt. 
rend., 200: 804>-5, Mar. 4, 1945. 

New Photoelectric-Grid Shutter Perfects 
Night Aerial Photography, Phot. Trade 
News, 9: 28-29, May 1945. (1 J500 sec.) 

The High-Speed Motion Picture Camera, H. 
J. Smith, Commercial Phot., 20: 248-51, 
May; 281-84, 285, June 1945. 

8000 Pictures per Second, H. J. Smith, Jour. 
SMPE, 45: 171-83, Sept. 1945. 

Two-Channel Ballistics Camera, N. E. 
Alexander, Office of Technical Services, P.B. 
L 58,200, Dec. 1945. 

Robatron Camera, A. Busch, Pop. Phot., 18: 
110, 112, #2, Feb. 1946. 

A Wide-Angle 35-Mm High-Speed Motion- 
Picture Camera, J. H. Waddell, Jour. 
SMPE, 46: 87-102, Feb. 1946; cf. Bell 
Lab. Record, 24: 139-14, Apr. 1946. 

The Scophony High-Speed Camera, Phot. J., 
86B: 42-46, #2, Mar.-Apr. 1946. 

High-Speed Camera, K. M. Baird, Can. J. 
Research, 24A: 41-45, July 1946. 

Wartime Camera Turns Civilian, M. Royer, 
Pop. Phot., 19: 82, 84, July 1946. (Fair- 
child K-25 Sequence camera.) 

High-Speed Camera, C. D. Miller, Mech. 
Eng., 68: 903, Oct. 1946. 

Optical Problems of the Image Formation in 
High-Speed Motion Picture Cameras, J. 
Kudar, Jour. SMPE, 47: 400-2, Nov. 
1946. 



January 1951 Journal of the SMPTE Vol. 56 



Pictures of the Invisible New Machine 
Takes Super-Speed Photos, U. S. Camera, 
9: 32, #10, Nov. 1946. 

Fastax High-Speed Camera: 4000 Pictures 
a Second, Amer. Cinemat., 28: 7,22, Jan. 
1947. 

New 16-Mm Professional Camera, F. F. 
Baker, Jour. SMPE, 48: 157-61, Feb. 
1947. 

The Marley High-Speed Camera, R. H. 
Bomback, Photography (New Series), 2: 
28, 32, May- June 1947. 

Photography at Five Million Frames per 
Second, B. O'Brien, Elec. Eng., 67: 157, 
Feb. 1948. 

U. S. Navy Magnified Time: Special High- 
Speed Cameras, Product Eng., 19: 147-48, 
Aug. 1948; cf. Electronics, 21: 164-66, 
July 1948. (Zarem camera, 400,000 frames 
per sec.) 

40,000 Frames per Second, C. D. Miller, 
PSA Jour., 14: 669-74, Nov. 1948. (De- 
tailed illustrations of the camera.} 

Child Photography, the Modern Way, J. 
Schneider, The Camera Mag., Baltimore, 
Md., 1949. 

Motion Picture Photography at Ten Million 
Frames per Second, B. O'Brien and G. 
Milne, Jour. SMPE, 52: 30^1, Jan. 
1949. 

Motion Picture Equipment for Very High- 
Speed Photography, B. O'Brien and G. G. 
Milne, Jour. SMPE, 52: 42-48, Mar. Pt. 2, 
1949. 

Lenses for High-Speed Motion Picture 
Cameras, A. A. Cook, Jour. SMPE, 52: 
110-15, Mar. Pt. 2, 1949. 

The Research Camera Greatly Extends 
Man's Vision. 288,000 Frames a Minute, 
Phot. Age, 4: 15-16, 19, June 1949. 

New High-Speed Camera, International 
Phot., 21: 18-19, Aug. 1949. 



High-Speed Recording by a Rotating-Mirror 
Method, J. D. Owen and R. M. Davies, 
Nature, 164: 752, Oct. 29, 1949. 

High-Speed Motion Pictures by Multiple- 
Aperture Focal-Plane Scanners, F. E. 
Tuttle, Jour. SMPE, 53: 451-61, Nov. 
1949. 

Improvements in High-Speed Motion Pic- 
tures by Multiple-Aperture Focal-Plane 
Scanners, F. E. Tuttle, Jour. SMPE, 63: 
462-68, Nov. 1949. 

Twenty-Lens High-Speed Camera, C. W. 
Wyckoff, Jour. SMPE, 53: 469-78, Nov. 
1949. 

Half-Million Stationary Images per Second 

With Refocused Revolving Beams, C. D. 

Miller, Jour. SMPE, 63: 479-88, Nov. 

1949. 
Very High-Speed Drum-Type Camera, K. 

M. Baird and D. S. L. Durie, Jour. SMPE, 

53: 489-95, Nov. 1949. 

Design of Rotating Prisms for High-Speed 
Cameras, J. H. Waddell, Jour. SMPE, 63: 
496-501, Nov. 1949. 

Bowen Ribbon-Frame Camera, E. E. Green 

and T. J. Obst, Jour. SMPE, 53: 515-23, 

Nov. 1949. 
Race Finish Recorder, J. C. Beckman and E. 

M. Whitley, Electronics, 22: 98-100, Dec. 

1949. 
High-Speed Synchroscope, G. G. Kelley, 

Rev. Sci. Instr., 21: 71-76, Jan. 1950. 

Hand-Held High Speed Motion Picture 
Camera, B. Marcus, Photographic Engi- 
neering, 1: 57-62, Apr. 1950. 

Streak Photography, I. Vigness and R. C. 
Nowak, J. Appl. Phys., 21: 445-48, May 
1950. 

A 100,000,000 Frame per Second Camera, M. 
Sultanoff, Rev. Sci. Instr., 21: 653, July 
1950. 



III. LIGHTING 



Sur la cinematographic ultrarapide, H. 

Abraham, E. Bloch and L. Bloch, Compt. 

rend., 169: 1031-33, Dec. 1, 1919. (Bull's 

oscillating spark light source.) 
Lighting for Color Photography, Gjon Mili, 

J. Phot. Soc. Amer., 4: 11-13, Summer 

1938. 
Speedlamp, M. L. Sandell, Complete Phot., 9: 

3281-88, Issue 51, 1943. 
10-Millionths-Second Exposures Used to 

Photograph Water Spray, Phot. Trade 

News, 7: 19, Mar. 1943; cf. Westinghouse 

Eng., 3: 50, May 1943. (1 {10,000,000 of a 

sec. high-intensity flash from a 5500-v spark 

gap.) 



Four-Microsecond Flash Unit, Electronics, 16: 

144, 146, Oct. 1943. 
Photoflash Lamps Motion Pictures, H. M. 

Lester, Gen. Elec. Rev., 47: 19-20, Apr. 

1944. 
Continuous Flash Lighting An Improved 

High-Intensity Light Source for High- 
Speed Motion Picture Photography, H. M. 

Lester, Jour. SMPE, 45: 358-69, Nov. 

1945. 
Speedlights, A. Palme, Amer. Phot. Publ. Co., 

Boston, 1946. 
Light Source for High-Speed Photography, 

Gen. Radio Co., Electronics, 19: 302, Jan. 

1946. 



Bibliography on High- Speed Photography 



97 



Speedlight and Photoflash, A. Palme, Amer. 
Phot., 40: 28-29, July 1946. 

Continuous-Flash Illumination for High- 
Speed Motion Picture Photography, H. 
M. Lester, J. Phot. Soc. Amer., 12: 623-29, 
Nov. 1946. 

Continuous Flash Illumination for High- 
Speed Motion Picture Photography, H. M. 
Lester, Phot. Age, 2: 22-26, 39, Jan. 
1947. 

Characteristics and Application of Flash- 
tubes, F. E. Carlson and D. A. Pritchard, 
Illuminating Eng., 42: 235-48, Feb. 1947. 

Flashtubes A Potential Illuminant for Mo- 
tion-Picture Photography, F. E. Carlson, 
Jour. SMPE, 48: 395-406, May 1947. 

Spark Light-Source of Short Duration, J. W. 
Beams, A. R. Kuhlthau, A. C. Lapsley, J. 
H. McQueen, L. B. Snoddy and W. D. 
Whitehead, Jr., J. Opt. Soc. Amer., 87: 
868-70, Oct. 1947. 

Discharge Lamps for Photography and Pro- 
jection, H. K. Bourne, Chapman, London, 
1948. 

The High Intensity Flash-Discharge Tube, 
J. N. Aldington, Endeavour, 7: 21-26, 
Jan. 1948. 

Microflash Unit for Ballistic Photography, 
W. W. McCormick, L. Madansky and A. 
F. Fairbanks, J. Appl. Phys., 19: 221-25, 
Mar. 1948. 

High-Speed Flash Unit, A. J. Drugan, 
U. S. Camera, 11: 22-23, June 1948. 

The Flash-tube and Its Applications, J. N. 
Aldington and A. J. Meadowcroft, J. 
Inst. Elec. Eng., 95: 671-81, Part 2, Dec. 
1948. 



Characteristic of Photoflash Lamps, M. P. 
Murgal, J. Sci. Ind. Res., 8: 44-50, Feb. 
1949. 

Lamps for High-Speed Photography, R. E. 
Farnham, Jour. SMPE, 52: 35-51, Mar. 
Pt. 2, 1949. 

New High-Speed Stroboscope for High- 
Speed Motion Pictures, K. J. Germeshau- 
sen, Jour. SMPE, 52: 24-34, Mar. Pt. 2, 
1949. 

High-Speed Flash for Black and White and 
Color, H. G. Morse, PSA Jour., 15: 
242-44, Apr. 1949. 

G. E. Develops Powerful Photo Lamp for 
High-Speed Film Production, Business 
Screen Mag., 10: 44, #2, Apr. 1949. 

High-Intensity Flash-Tubes, G. Knott, 
Phot. J., 89B: 46-50, May-June 1949. 

High-Speed Flash Tubes and Their Applica- 
tions, A. J. Meadowcroft, Phot. J., 89B: 
51-53, May-June 1949. 

A High-Intensity Light Source for High- 
Speed Kinematography, E. J. G. Beeson, 
Phot. J., 89B: 62-67, May-June 1949; 
cf. Nature, 164: 453-54, Sept. 10, 1949. 

New Lighting for High-Speed Photography, 
W. R. Plant, Gen. Elec. Rev., 52: 22-27, 
June 1949; cf. Phot. Age, 4: 10-11, 22-23, 
Sept. 1949; 8-10, Oct. 1949. 

Tracer Lights, Modern Photography, 13: 
44-47, 126-27, Oct. 1949. 

Cine-Flash Equipment for High-Speed Cine- 
photography, H. K. Bourne, Functional 
Photography, 1: 17-19, Jan. 1950. 

The Stroboscope as a Light Source for Mo- 
tion Pictures, R. S. Carlson and H. E. 
Edgerton, Jour. SMPTE, 55: 88-100, 
July 1950. 



IV. OSCILLOGRAPHY 



Improved Cathode-Ray Tube Method of the 
Harmonic Comparison of Frequency, D. 
W. Dye, Proc. Phys. Soc. (London), 37: 
158-68, Apr. 1925. 

A Photokymographic Method with Continu- 
ous Cathode Ray Oscillograms, W. S. 
McCulloch and G. R. Wendt, Science, 83: 
354-55, Apr. 10, 1936. 

New Siemens and Halske Cathode Ray 
Oscillograph, W. Gaarz and P. P. Klein, 
Siemens Rev., 13: 138-48, #4, 1937. 

"fiber Kathodenstrahlphotographie, W. 
Holzer, Phot. Korr., 73: 9-14, Jan. 1937. 

A Film Camera for Use with the Cathode- 
Ray Oscillograph, Specially Designed for 
Experiments on Artificial Radioactivity, 
L. G. Grimmett and W. H. Rann, J. Sci. 
Instr., 14: 96-100, Mar. 1937. 



The Recording of Rapidly Occurring Elec- 
tric Phenomena With the Aid of the 
Cathode-Ray Tube and the Camera, J. F. 
H. Custers, Philips Tech. Rev., 2: 148-55, 
May 1937. 

Nachbeschleunigungs Elektronstrahl Ozil- 
lograph, A. Bigalke, Zeit. tech. Physik, 19: 
163-66, #6, 1938. 

Oscillograph Film Camera for Brief Records, 
A. J. Small, J. Sci. Instr., 15: 238-39, 
1938. 

Schaltvorrichtung fur Aufnahmen kurz- 
zeitiger elektrischer Vorgange mit dem 
Siemens-Schliefen Ozillographen, W. Neu- 
mann, Zeit. Instrumentenkunde, 58: 41-43, 
Jan. 1938. 



98 



January 1951 Journal of the SMPTE Vol. 56 



A Recurrent Surge Oscillograph, G. J. 
Scoles, J. Sci. Instr., 15: 201-5, June 
1938. 

An Apparatus for the Measurement of 
Scanning Speeds of Cathode-Ray Tubes, L. 
Blok, Philips Tech. Rev., 3: 216-19, July 
1938. (Apparatus is described for the deter- 
mination of the greatest scanning speed of 
cathode-ray tubes which can be photographed. ) 

Use of High-Vacuum Cathode-Ray Tube 
for Recording High-Speed Transient Phe- 
nomena, D. I. McGillewie, J. Inst. Elec. 
Eng., 83: 657-80, Nov. 1938; Elec. Com- 
munication, 17: 124-32, Oct. 1938. 

Principles et construction des oscillographes 
a rayons cathodiques, M. Demontvignier, 
Rev. gen. elec., 45: 3-14, Jan. 7, 1939; 
38-42, Jan. 14, 1939. 

Oscillographe cathodique simple pour 1'enre- 
gistrement de phenomenes transitoires, J. 
Cuilhe and T. Vogel, Rev. gen. elec., 45: 
103-6, Jan. 28, 1939. 

The Testing of Electric Fuses with the 
Cathode-Ray Oscillograph, J. A. M. 
Liempt and J. A. deVriend, Philips Tech. 
Rev., 4: 118-20, Apr. 1939. 

Cathode-Ray Screen Photography, Elec- 
tronics, 11: 37-38, Apr. 1939. 

Cathode-Ray Tube Photography, T. A. 
Rogers and B. L. Robertson, Electronics, 
12: 19, July 1939. 

A Cathode-Ray Oscillograph, J. D. Veegens, 
Philips Tech. Rev., 4: 198-204, July 1939. 
(Possibility of observing or photographing 
single phenomena of short duration is dis- 
cussed.) 

Cathode-Ray Photography, J. H. Jupe, 
Wireless World, 45: 233-34, Sept. 7, 1939. 
(Choice of films and plates, exposure, aper- 
ture and reduction of image, formulas for de- 
velopers.) 

A New Oscillograph for the Testing of Shot- 
Firing Machines (in German), W. Hartel, 
Siemens Zeit., 499-508, Nov. 1939. (Oscil- 
lograph is of electromagnetic type with photo- 
graphic recording on a drum driven at 1500 
rpm.) 

Automatic Cathode-Ray Oscillograph, W. 
L. Gaines, Bell Lab. Record, 18: 145-50, 
Jan. 1940. (The photographic mechanism 
is described.) 

Testing Amplifier Output Valves by Means 
of the Cathode-Ray Tube, A. J. Heins van 
der Ven, Philips Tech. Rev., 5: 61-68, Mar. 
1940. 

Ueber die Beurteilung und den objektiven 
Vergleich der Messleistung von Katho- 
denstrahloszillographen, B. von Borries 
and E. Ruska, Arch, fur Elektrotech., 34: 
161-66, Mar. 1940. (The output of a 



cathode-ray oscillograph as a measuring in- 
strument is determined by the speed of the 
spot and the sensitivity. The speed is the 
maximum at which an impression can be 
made on a photographic plate; this speed 
combined with the diameter of the spot is a 
measure of the resolving power. The sensi- 
tivity combined with the spot diameter mea- 
sures the accuracy. In practice, a compro- 
mise between the two qualities is the most 
satisfactory solution and certain rules are 
proposed.) 

A Recording System Designed for the In- 
vestigation of the Electrical Relations in 
the Brains of Small Animals, G. A. Woon- 
ton, Can. J. Research, ISA: 65-73, Apr. 
1940. (Two RCA high-vacuum cathode-ray 
tubes and a motor-driven Stoppani camera 
were used.) 

Cathode-Ray Oscillograph for Use in Tool 
Making, S. L. de Bruin and C. Dbrsman, 
Philips Tech. Rev., 5: 277-85, Oct. 1940. 

A High- Voltage Cathode-Ray Oscillograph, 
A. C. Hall and J. M. Coombs, Rev. Sci. 
Instr., 11: 314-20, Oct. 1940. 

Photography of Cathode-Ray Tube Traces, 
H. F. Folkerts and P. A. Richards, RCA 
Rev., 6: 234-44, Oct. 1941. 

A Cathode-Ray Oscillograph with Rotating 
Drum Camera, E. G. Downie, Trans. 
Elec. Eng., 60: 984-85, Nov. 1941. 

Die Verwendung von Kaltkathodenstrahlos- 
zillographen, G. Induni, Schweiz. Arch- 
angew. wiss. Tech., 8: 35-45, Feb. 1942. 
(Examples are given of the use of single- and 
double-ray cold-cathode c.r. oscillographs for 
electron refrat tion investigation of molecular 
structure with numerous photographs.) 

Errors in Photography of Cathode-Ray Tube 
Traces. The Effects of Screen Curvature, 
H. Moss and E. Cattanes, Electronic Eng. 
14: 720-21, Apr. 1942. 

The Production of Elastic Waves by Explo- 
sion Pressures, II., J. A. Sharpe, Geo- 
physics, 7: 311-21, July 1942. (A cathode- 
ray oscilloscope was used with ultraspeed 
film photosensitized with mercury.) 

The Cathode-Ray Tube Used Stroboscopi- 
cally, G. Bocking, Electronic Eng., 15- 
102-3, Aug. 1942. 

The New Fastax High-Speed Camera, C. L. 
Strong, Amer. Cinemat., 24: 292-93, 297, 
Aug. 1943; cf. H. J. Smith, Bell Lab. 
Record, 22: 1-4, #1, Sept. 1942. 

Application of Oscillograph to Determination 
of Cooling Rates of Quenched Steels, C. R. 
Austin, R. M. Allen and W. G. VanNote, 
Trans. Amer. Soc. Metals, 30: 747-75, 
Sept. 1942. 



Bibliography on High-Speed Photography 



99 



Applications of Cathode-Ray Tubes, B. 

Dudley, Electronics, 15: 49-52, 154-55, 

Oct. 1942. 
Recording Machinery Noise Characteristics, 

H. D. Brailsford, Electronics, 15: 46-51, 

164-66, Nov. 1942. (Includes description of 

apparatus.) 
A Concise Report on a High-Speed Camera 

for Simultaneous Photographic and Oscil- 

lographic Records, H. W. Baxter, J. Sci. 

Instr., 19: 183-84, Dec. 1942. 

The Photography of High-Speed Transient 
Phenomena with the Sealed-Off Glass- 
Tube Cathode-Ray Oscillograph, W. Neth- 
ercot, /. Sci. Instr., 20: 75-78, May 1943. 

Precision Stroboscopic Frequency Meter, 
E. L. Kent, Electronics, 16: 120-21, Sept. 
1943. 

Photography of Cathode-Ray Tube Traces, 
N. Hendry, Electronic Eng., 16: 324-26, 
Jan. 1944. 

Photographing Patterns on Cathode-Ray 
Tubes, R. Feldt, Electronics, 17: 130-37, 
262, 264, 266, Feb. 1944. 

Recording High-Speed Transient Phenomena 
by the Hot-Cathode, Glass-Bulb, Cathode- 
Ray Oscillograph, W. Nethercot, Electronic 
Eng., 16: 369-71, Feb. 1944; note on 
above by H. Moss, p. 411, Mar. 1944. 

A Continuous Film-Recording Camera for 
Use with Standard Cathode-Ray Oscillo- 
scopes, A. H. Simons, Electronic Eng., 18: 
10-12, Jan. 1946. 

Chemical Propellants. The System Hydro- 
gen Peroxide Permanganate, F. Bellinger, 
H. B. Friedman, W. H. Bauer, J. W. 
Eastes, J. H. Ladd and J. E. Ross, Ind. 
Eng. Chem., 38: 160-69, Feb. 1946. (A 
General Motors capacitance gage cathode-ray 
oscillograph 36-mm camera oscillograph re- 
corder was used to record pressure in chamber 
of a thrust motor.) 

High-Speed Photography of the Cathode- 
Ray Tube, H. Goldstein and P. D. Bales, 
Rev. Sci. Instr., 17: 89-96, Mar. 1946. 

High-Speed Oscillograph, N. Rohats, Elec- 
tronics, 19: 135-37, Apr. 1946. 

The Photographic Recording of Cathode- 
Ray Tube Screen Traces. The Choice of 
Emulsions and Developers, R. J. Hercock, 
Phot. J., 86B: 138-42, Nov.-Dec. 1946. 

The Photography of High-Speed Traces with 
the Sealed-Off Glass Cathode-Ray Tube, 



W. Nethercot, Phot. J., 86B: 159-64, 
Nov.-Dec. 1946. 

The Photographic Recording of Cathode-Ray 
Tube Traces, R. J. Hercock, Ilford Ltd., 
London, 1947. 

Photographing Cooling Curves of Hardening 
Oils by Means of a Cathode-Ray Oscillo- 
graph, Philips Tech. Rev., 9: 147-48, #5, 
1947. 

Cathode-Ray Tube Shutter-Testing In- 
strument, D. T. R. Dighton and H. McG. 
Ross, J. Sci. Instr., 24: 128-33, May 
1947. 

Recording Oscilloscope Images, Tele-Tech, 6: 
45-46, Apr. 1947; cf. Instruments, 20: 1052, 
Nov. 1947. 

Note on the Photography of Cathode-Ray 
Oscillograms, M. V. Scherb, Rev. Sci. Instr., 
18: 925, Dec. 1947. 

Multi-Channel Recording of Oscillograph 
Traces, Engineer, 184: 560, Dec. 12, 1947. 

An Oscilloscope Camera, H. E. Hale and H. 
P. Mansberg, Electronics, 21: 102-7, 
June 1948; cf. Instruments, 20: 1052-53, 
Nov. 1947. (1-3600 feet per minute.) 

Illumination of Cathode-Ray Oscillograph 
Screen for Photography, E. C. Crittenden, 
Jr., C. S. Smith and L. O. Olsen, Rev. Sci. 
Instr., 19: 271-72, Apr. 1948. 

Continuous Oscillograph Camera for Iono- 
sphere Measurements, J. E. Hacke, Jr., 
Instruments, 21: 914-15, Oct. 1948. 

Modern Oscilloscopes and Their Uses, J. H. 
Ruiter, Jr., Murray Hill Books, New 
York, 1949. 

Oscillograph for Automatic Recording of 
Disturbances on Electric Supply Systems, 
W. T. J. Atkins, J. Inst. Elec. Eng., 96: 
276-85, Pt. 2, Apr. 1949. 

Cathode-Ray-Tube Applications in Pho- 
tography and Optics, C. Berkley and R. 
Feldt, Jour. SMPE, 52: 64-85, July 1949; 
Electronic Eng., 21: 114-20, Apr. 1949; 
169-72, May 1949. 

A Miniature Portable Cathode-Ray Oscillo- 
graph Recorder, C. F. Johnson, Instru- 
ments, 22: 800-1, Sept. 1949. 

Techniques in High-Speed Cathode-Ray 
Oscillography, C. Berkley and H. P. 
Mansberg, Jour. SMPE, 63: 549-78, Nov. 
1949. 

Improved C-R Photographs, N. Fulmer, 
Electronics, 23: 86-87, Mar. 1950. 



100 



January 1951 Journal of the SMPTE Vol. 56 



V. SCHLIEREN 



Photography of Flying Bullets, etc., C. V. 
Boys, Phot. J., 16: 199-209, Apr. 30, 1892; 
cf. Nature, 47: 415-21, Mar. 2, 1893; 
440-46, Mar. 9, 1893; Smithsonian Inst. 
Report for 1892-98, 165-82. 

Photography of Sound Waves by the "Schlie- 
ren Methode," R. W. Wood, Phil. Mag., 
(5), 48: 218-27, Aug. 1899. 

Photography of Sound Waves, R. W. Wood, 
Phot. J., 24: 250-56, May 31, 1900. 

The Photography of Sound Waves and the 
Demonstration of the Evolutions of Re- 
flected Wave Fronts with the Cinemato- 
graph, R. W. Wood, Phil. Mag. (5), 60: 
148-56, 1900; cf. Nature, 62: 342-49, 
Aug. 9, 1900; Smithsonian Inst. Report for 
1900, 359 ff., 1901; Proc. Roy Soc., 66: 
283-90, #429, 1900. 

Movements of Flame in Explosions of Gases, 
H. B. Dixon, Phil. Trans. Roy. Soc., 200A: 
315-52, Feb. 4, 1903. 

Beobachtungen nach der Schlierenmethode, A. 
Toepler, W. Engelmann, Leipzig, 1906. 

Photographing Air Waves and Bullets, W. 
Hyde, Brit. J. Phot., 63: 492-93, Sept. 8, 
1916. 

tiber Hochfrequenz - Schlierenkinemato- 
graphie und ihre Verwendung zur Unter- 
suchung von Explosionserscheinungen und 
anderen sehr rasch verlaufenden Vor- 
gangen, C. Cranz and E. Barnes, Z. angew. 
Chem., 36: 76-80, Feb. 7, 1923. 

tiber Hochfrequenzschlierenphotographie, E. 
Barnes, Phot. Rund., 61: 120-23, July 
1924. 

Die Photographie von Explosionsvorgangen, 
H. Moss, Phot. Korr., 66: 69-73, #3, 1930. 

Schlieren Kinematographie, P. Schrott, 
Kinotechnik, 12: 40-46, Jan. 1930. 

Mehrfachfunkenaufnahmen von Explosions- 
vorgangen nach der Toeplerchen Schlie- 
renmethode, W. Lindner, Kinotechnik, 12: 
356-60, July 1930. 

Applications of the Schlieren Method of 
Photography, D. B. Gawthrop, Rev. Sci. 
Instr., 2: 522-31, Sept. 1931. 

Ignition of Firedamp by Explosives, Study 
of the Process of Ignition by the Schlieren 
Method, W. C. F. Shepherd, Bull, of the 
Bureau of Mines, #354, 1932. 

Improvements in the Schlieren Method, H. 
G. Taylor and J. M. Waldram, J. Sci. 
Instr., 10: 378-89, Dec. 1933. 

Sichtbarmachung von Schwingungen einer 
Quarzplatte mittels der Schlierenmethode, 
V. Petrzilka and L. Zachoval, Zeit. Physik, 
90: 700-02, #9/10, 1934. 



Toepler's Schlieren Method. Fundamentals 
of Its Application and Quantitative Use, 
H. Schardin, V. D. I. Forschungsheft, 
Berlin, #367, 32 pp., 1934. (Translated by 
I.E. Segal, R TP Translation #249. Bibliog- 
raphy.) 

Improvements in the Schlieren Method of 
Photography, H. C. H. Townend, J. Sci. 
Instr., 11: 184-87, June 1934. 

Die Harze der Steinkohlen, H. Winter, 
Angew. Chem., 48: 610-14, Sept. 21, 1935. 
(Schlieren photographs indicate that fossil 
resins are high molecular esters.) 

Warmubergang durch freie Konvektion an 
Quadratischen Flatten, R. Weise, Forsch. 
Gebiete Ingenieurw., 6: 281-92, Nov.-Dec. 
1935. (Photographs taken by Schlieren 
method of temperature fields.) 

Observation of Ultracentrifugal Sedimenta- 
tion by the Toepler-Schlieren Method, A. 
Tiselius, K. O. Pedersen and I. Eriksson- 
Quensel, Nature, 139: 546, Mar. 27, 1937. 

Measurement of Heat Transfer by Free 
Convection from Cylindrical Bodies by 
Schlieren Method, L. M. K. Boelter and 
V. H. Cherry, Trans. Amer. Soc. Heating 
and Ventilating Engrs., 44: 499-512, 1938. 

Photographing the Invisible, J. H. W. 
Kerston, Pop. Phot., 6: 16-17, 74, 76, Apr. 
1940. (Schlieren photography.) 

Schlierenaufnahmen von Gasflammen, A. 
N. J. van de Poll and T. Westerdijk, Zeit. 
tech. Physik, 22: 29-32, #2, 1941. 

Schlieren and Shadowgraph Equipment for 
Air Flow Analysis, N. F. Barnes and S. L. 
Bellinger, J. Opt. Soc. Amer., 35: 497-509, 
Aug. 1945. 

High-Speed Schlieren Camera for Observa- 
tion of Flame Travel,' BIOS Report #14, 
Item 9; P. B. Report L85, 201, 1947. 

Knocking Combustion Observed in a Spark- 
Ignition Engine with Direct and Schlieren 
High-Speed Motion Pictures and Pressure 
Records, G. E. Osterstrom, National Ad- 
visory Committee for Aeronautics, Tech. 
note #1614, June 1948. 

Electron Optical Schlieren Effect, Tech. News 
Bull. Nat. Bur. Stand., 32: 82-84, July 
1948. 

The Operation of the Puff-Ball Mechanism 
of Lycoperdon Perlatum by Raindrops, 
Shown by Ultra-High-Speed Schlieren 
Cinematography, P. H. Gregory, Trans. 
British Mycological Soc., 32: 11-15, Pt. 1, 
1949. 



Bibliography on High-Speed Photography 



101 



The Application of the Schlieren Method to 
the Quantitative Measurement of Mixing 
Gases in Jets, W. R. Keagy, Jr. and H. H. 
Ellis, pp. 667-74, in: Third Symposium on 
Combustion and Flame and Explosion 
Phenomena, Williams and Wilkins, Balti- 
more, Md., 1949. 

Visual Methods for Studying Ultrasonic 
Phenomena, R. B. Barnes and C. J. 
Burton, J. Appl. Phys., 20: 286-96, Apr. 
1949. (Modifications of Schlieren photog- 
raphy.) 



Interference Phenomenon in the Schlieren 
System, E. L. Gayhart and R. Prescott, 
J. Opt. Soc. Amer., 39: 546-50, July 
1949. 

Physical Optic Analysis of Image Quality in 
Schlieren Photography, H. J. Shafer, 
Jour. SMPE, S3: 524-44, Nov. 1949. 

A Modified Schlieren Apparatus for Large 
Areas of Field, R. A. Burton, J. Opt. Soc. 
Amer., 39: 907-8, Nov. 1949. 

Improved Schlieren Apparatus Employing 
Multiple Slit-Gratings, T. A. Mortensen, 
Rev. Sci. Instr., 21: 3-6, Jan. 1950. 



VI. TECHNICAL AND TECHNIQUES 



Photography of Flying Bullets, etc., C. V. 

Boys, Phot. J., 16: 199-209, Apr. 30, 

1892; cf. Nature, 47: 415-21, Mar. 2, 

1893; 440-46, Mar. 9, 1893; Smithsonian 

Inst. Report for 1892-93, 165-82. 
Photography of Sound Waves, R. W. Wood, 

Phot. J., 24: 250-56, May 31, 1900. 
A Study of Splashes, A. M. Worthington, 

Longmans, New York, 1908. 
Photographing Air Waves and Bullets, W. 

Hyde, Brit. J. Phot., 63: 492-93, Sept. 8, 

1916. 
De la chrono-photographie par etincelle 

electrique & 1'etude des phenomenes de 

ballistique, L. Bull, Bull. rech. et inv., 1: 

48-54, Nov. 1919. 
The Spectrum of Electrically Exploded 

Wires, J. A. Anderson, Astrophys. J., 51: 

37^8, Jan. 1920. 
Appareil pour la dissociation rapide des 

images dans le cinematographic par 

etincelle electrique, L. Bull, Compt. rend., 

174: 1059-1061, Apr. 18, 1922. 
Photography of Bullets in Flight, P. P. 

Quayle, J. Franklin Inst., 193: 627-40, 

May 1922. 
High Speed Photography of Vibrations 

(Sound, Mechanical, Electrical, etc.), A. 

Trowbridge, /. Franklin Inst., 194: 713-29, 

Dec. 1922. 

A Photographic Study of High Voltage Dis- 
charges, R. H. George, K. B. McEachron 

and K. A. Oplinger, Bull. Purdue Univ. 

Eng. Expt. Station #19, 118 pp., 1924. 
The Movement of Flame in Closed Vessels, 

O. C. deC. Ellis and R. V. Wheeler, J. 

Chem. Soc., 127: 764-67, 1925. 
Spark Photography and Its Application to 

Some Problems in Ballistics, P. P. Quayle, 

Sci. Papers Bur. Stand. #508, 20: 237-76, 

1925. 



Single Spark Photography and Its Applica- 
tion to Some Problems in Ballistics, P. P. 
Quayle, Nature, 115: 765-70, May 16, 
1925. 

Spark Photography as a Means of Measuring 
Rate of Explosion, J. E. Smith, Phys. Rev., 
26: 870-76, June 1925. 

Tagging a Bullet on the Wing, S. P. Meek, 
Sci. Amer., 81: 158-59, Sept. 1925. (One 
millionth of a second.) 

Application de la photographic sur plaque 
mobile a 1'etude du mouvement des pro- 
jectiles et en particulier la mesure de 
leur vitesse, G. Foex and J. Kampe de 
Feriet, Compt. rend., 181: 597-99, Nov. 3, 
1925. 

General Characteristics of Electrically Ex- 
ploded Wires, J. A. Anderson and S. 
Smith, Astrophys. J., 64: 295-314, Dec. 
1925. 

The Pressure Wave Caused by an Explosion, 
W. Payman and H. Robinson, Safety in 
Mines Research Board Paper, #18, 1926; 
W. Payman and W. C. F. Shepherd, #29, 
1926. 

Lehrbuch der Ballistik, Band 3, C. Cranz, 
Springer, Berlin, 1927. 

Photographic Measurement of Rate of De- 
tonation of Explosives, G. St. J. Perrott 
and D. B. Gawthrop, J. Franklin Inst., 
203: 103-10, Jan. 1927; 387-406, Mar. 
1927. 

Flame, O. C. deC. Ellis, Phot. J., 67: 349-58, 
Aug. 1927. 

Cinematographic Studies in Aerodynamics, 
A. Klemin, Mech. Eng., 50: 217-220, 
Mar. 1928. 

Some Applications of Spark Photography, 
D. B. Woodbridge and A. E. Palmer, J. 
Opt. Soc. Amer., 16: 125-33, #2, 1928. 



102 



January 1951 Journal of the SMPTE Vol. 56 



Ein neuer Trommelaufnahme Apparat, F. 
Beck, Kinotechnik, 11: 124-26, Mar. 5, 
1929; cf. Lichtbildbuhne, 22: 18-19, Apr. 
13, 1929. 

Photographic Investigation of Flame Move- 
ments in Carbonic Oxide-Oxygen Explo- 
sions, W. A. Bone and R. P. Frazer, Phil. 
Trans. Roy. Soc., 228: 197-234, May 17, 
1929. 

Photographing Sound, C. Neblette, Photo 
Era, 64: 160-63, #3, 1930. (Foley's ap- 
paratus for spark photography is illus- 
trated.) 

Les ondes stationnaires ultra-sonores rendues 
visibles dans les gaz par la methode des 
stries, E. P. Tawil, Compt. rend., 191: 
92-95, July 16, 1930. 

Application of Photography in Explosives 
Research, W. Payman, Phot. J., 70: 409- 
11, 419, Sept. 1930. 

Methode d'observation d'ondes sonores non 
stationnaires, E. P. Tawil, Compt. rend., 
191: 998-1000, Nov. 24, 1930. 

The Photography of Waves and Vortices 
Produced by the Discharge of an Explo- 
sive, D. B. Gawthrop, W. C. F. Shepherd 
and G. St. J. Perrott, J. Franklin Inst., 
211: 67-86, Jan. 1931. 

The Propagation of Flame in Electric Fields. 
I. Distortion of the Flame Surface, E. M. 
Guenault and R. V. Wheeler, /. Chem. 
Soc., 195-99, Jan. 1931. 

Explosions in Closed Cylinders. IV. Correla- 
tions of Flame Movement and Pressure 
Development in Methane-Air Explosions, 
W. A. Kirkby and R. V. Wheeler, J. 
Chem. Soc., 847-53, Apr. 1931; V. The 
Effect of Restrictions, 2303-6, 1931. 

Zeithehner-Untersuchungen, W. Ende, 
Kinotechnik, 13: 139-42, Apr. 20, 1931; 
158-61, May 5, 1931. 

Cinematographic Analysis of Mechanical 
Energy Expenditure in the Sprinter, C. A. 
Morrison and W. O. Fenn, Jour. SMPE, 
16: 603-11, May 1931. 

Sur un cinematographic ultra-rapide donnant 
de 2000 a 3000 images par seconde, E. 
Hugenard and A. Magnan, Compt. rend., 
192: 1370-72, June 1, 1931. (Four lenses 
taking 12 pictures, in area, the size of one 
standard 35-mm frame.) 

Die Entwicklung der Mikrokinemato- 
graphie, H. Linke, Filmtechnik, 7: 1-4, 
June 27, 1931. 

Explosion Waves and Shock Waves, Pt. 1. 
The Wave-Speed Camera and Its Applica- 
tion to the Photography of Bullets in 
Flight, W. Payman and D. W. Woodhead, 
Proc. Roy. Soc., 132A: 200-13, July 2, 
1931. 



Applications de la photographic strobo- 
scopique a 1'etude des moteurs a huile 
lourde, Clerget, Bull. soc. francs, phot. (3), 
18: 180-83, Sept. 1931. 

Iowa Eye-movement Camera, H. H. Jasper 
and R. Y. Walker, Science, 74: 291-94, 
Sept. 18, 1931. 

Der Lidschluss in der Zeitlupenaufnahme, H. 
Pander, Kinotechnik, 13: 366-67, Oct. 5, 
1931. 

Neue Ergebnisse der Hochfrequenz-Kinema- 
tographie in der technischen Forschung 
(mit Filmvorf tihrung) , W. Ende, Ber. 
Int. Kong. Phot. Dresden, 1931, 265-71, 
1932. 

Ignition of Firedamp by Explosives, Study of 
the Process of Ignition by the Schlieren 
Method, W. C. F. Shepherd, Bull, of the 
Bureau of Mines, #354, 1932. 
Die Grenzen der Hochfrequenzkinemato- 
graphie, H. Schardin, Kinotechnik, 14 : 
41-45, Feb. 5, 1932. (Spark illumination is 
3 million per sec.) 

The Meteorology of Gas Explosions, O. C. 
deC. Ellis, Phot. J., 72: 380-84, Sept. 
1932. 

New Ways of Splitting Seconds, C. H. 
Fetter, Bell Telephone Quarterly, 11: 293- 
300, Oct. 1932. 

Propagation Tests and the Photography of 
the Disturbance Sent Out by the Explosion 
of Commercial Electric Detonators, D. B. 
Gawthrop, /. Franklin Inst., 214:- 647-64, 
Dec. 1932. 

On the Propagation of a Lightning Dis- 
charge through the Atmosphere, E. C. 
Halliday, Phil. Mag. (7), 15: 409-20, 
Feb. 1933. 

La cinematographic par etincelles, L. Bull, 
Reunions de VInstitut d'Optique, 5: 13-22, 
1934. 

Utilisation de 1'etincelle electrique en balis- 

tique et eri aerodynamique, L. Libesscut, 

Reunions de VInstitut d'Optique, 5: 5-12, 

1934. 
Photo-Record of the Speed of an Explosion 

Wave in an Electric Field (in English), A. 

E. Malijowski, B. N. Naugolnikow and K. 

T. Tkatschenko, Physik. Zeit. Sowjet- 

union, 5: 446-52, #3, 1934. 

The Pressure Wave Sent Out by an Ex- 
plosive. III. Spark Photographs With 
Permitted Explosives, W. Payman and D. 
W. Woodhead, Safety in Mines Res. Board 
Paper #88, 1934. 

Determination d'eclairs rapides dans les 
tubes a gaz par les appareils "Strobo- 
rama," L. and A. Sequin, Reunions de VIn- 
stitut d'Optique, 5: 23-45, 1934. 



Bibliography on High-Speed Photography 



103 



Statistical Measurements of Turbulence in 
the Flow of Air through a Pipe, H. C. H. 
Townend, Proc. Roy. Soc., 146A: 180-211, 
#854, 1934. 

Lightning Camera Proves Life-Saver, S. J. 
Cox, Photography, 2: 30, Jan. 1934. 

A Photoelectric Cell Method of Measuring 
the Velocity of Projectiles, D. C. Rose, 
Can. J. Research, 10: 571-87, May 1934. 

The Use of the Thyratron as a Light Source 
for Recording Purposes, D. C. Rose, Can. 
J. Research, 10: 588-90, May 1934. 

Photography of Flame in Gaseous Explo- 
sions, R. P. Fraser, Phot. J., 74: 388-405, 
Aug. 1934. 

Speed Copes With Speed in Study of Design, 
F. A. Ramsdell and A. Palmer, Machine 
Design, 7: 13-17, Mar. 1935. (Analyzes 
motion of a loom.) 

New Developments in Micro Motion Picture 
Technic, H. Roger, Jour. SMPE, 24: 
475-86, June 1935. 

Fortschritte auf dem Gebiet der Hochfre- 
quenz Kinematographie, C. Cranz and H. 
Schardin, Zeit. Ver. dent. Ing., 79: 1075-79, 
Sept. 7, 1935. 

Progressive Lightning, Part I, B. J. F. 
Schonland and H. Collens, Proc. Roy. Soc., 
143 A: 654-74, Feb. 1934; Part II, B. J. 
F. Schonland, D. J. Malan and H. Collens, 
152A: 595-625, Nov. 15, 1935. 

liber den Verdampfungsvorgang nach kine- 
matographischen Aufnahmen an Dampf- 
blasen, W. Fritz and W. Ende, Physik. 
Zeit., 37: 391-401, #11, 1936. (Modifica- 
tion of Thun's high speed camera; process 
of evaporation shown by cinematograph 
pictures of bubbles of steam.) 

Effect of Nozzle Design on Fuel Spray and 
Flame Formation in High-Speed Com- 
pression-Ignition Engine, A. M. Rothrock 
and C. D. Waldron, Nat. Adv. Comm. for 
Aeronautics #561, 1936. 

How Does a Welding Electrode Fuse, J. 
Sack, Philips Tech. Rev., 1: 26-29, Jan. 
1936. 

Olympic Games by Stereoscopic Films, 
Kinemat. Weekly, 227: 54, Jan. 23, 1936. 

Photo-Elastic Study of Stresses Due to Im- 
pact, Z. Tuzi and M. Nisida, Phil. Mag. 
(7), 21: 448-73, #140, Feb. (Suppl.) 1936. 
(1 146,000 sec.) 

Ultra-High-Speed Motion-Picture Photog- 
raphy, Ind. Eng. Chem., News ed., 14: 
61-62, Feb. 10, 1936. (3,000 frames per 
sec.) 

Nouveau dispositif cinematographique pour 
I'enregistrement de phenomenes tres rapi- 
des, L. Bull and P. Girard, Compt. rend., 
202: 554-55, #7, Feb. 17, 1936. 



Nouvel appareil a plaque fixe pour 1'analyse 
cinematographiques des mouvements ra- 
pides, L. Bull, Bull. soc. francs, phot. (3), 
23: 105-8, Apr. 1936. 

High-Speed Motion Pictures in Industrial 
Development, H. I. Day, J. Brit. Inst. 
Cinematography, 4- 15, 25, Apr. 1936. 

High-Speed Motion Pictures of Engine 
Flames, G. M. Rassweiler and L. Withrow, 
Ind. Eng. Chem., 28: 672-77, June 1936. 

A Method of Air Flow Cinematography Ca- 
pable of Quantitative Analysis, H. C. H. 
Townend, J. Aeronautical Sci., 3: 343-53, 
Aug. 1936. 

Les applications de la photographie et de la 
cinematographic par etincelles a 1'etude 
des phenomenes balistiques et aerodyna- 
miques: Photographies et cinemato- 
graphies ultrarapides, P. Libessart, Bull, 
soc. franqs. phot. (3), 23: 192-204, Sept. 
1936. 

Photographic "Judge" at the Olympic 
Games, Brit. J. Phot., 83: 688-89, Oct. 
30, 1936. (Zeiss-Ikon duplicate stereocine- 
matograph apparatus.) 

Use of Motion Pictures in an Accurate Sys- 
tem for Timing and Judging Horse Races, 
H. I. Day, Jour. SMPE, 27: 513-28, Nov. 
1936. 

Sound Wave Stroboscope, B. F. McNamee, 

Electronics, 9: 24-26, 34, Nov. 1936. 
Ultra-Slow Motion Photography as Applied 

to Chemical Engineering Studies, G. J. 

Esselen and J. G. Hildebrand, Trans. 

Amer. Inst. Chem. Eng., 32: 557-65, Dec. 

25, 1936. 

High-Speed Cameras for Measuring the Rate 

of Detonation in Solid Explosives, W. 

Payman, W. C. F. Shepherd and D. W. 

Woodhead, Safety in Mines Research 

Board Paper #99, 29 pp., 1937. 
Welding & Welding Rods, J. Sack, Philips 

Tech. Rev., 2: 129-35, May 1937. 
Application de la photographie a 1'etude des 

luminosites produites par la detonation des 

explosifs, A. Michel-Levy, Bull. soc. 

francs, phot. (3), 24: 125-32, July 1937. 
High-Speed Camera for Propeller Research, 

E. L. Gayheart, Engineer, 164: 23-24, 

July 2, 1937. 
A Photographic Study of the Vacuum Spark 

Discharge, J. A. Chiles, Jr., J. Appl. 

Phys., 8: 622-26, Sept. 1937. 
Studies in Drop Formation as Revealed by 

the High-Speed Motion Camera, H. E. 

Edgerton, E. A. Hauser and W. B. Tucker, 

J. Phys. Chem., 41: 1017-28, Oct. 1937. 
Filming Operation on Blood Corpuscles, S. 

H. Kahn, Photography, 6: 45, Oct. 1937. 



104 



January 1951 Journal of the SMPTE Vol. 56 






Neuere Ergebnisse der Funkenkinemato- 
graphie, H. Schardin and W. Struth, 
Zeit. tech. Physik, 18: 474-77, Dec. 1937. 

A Photographic Study of Combustion and 
Knock in a Spark-Ignition Engine, A. M. 
Rothrock and R. C. Spencer, National Ad- 
visory Committee for Aeronautics Report 
#622, Laboratories, Langley Field, Va., 
21 pp., 1938. 

High-Speed Motion Picture Photography 
Applied to Design of Telephone Apparatus, 
W. Herriott, Jour. SMPE, 80: 30-37, Jan. 
1938. 

Cinema Section: The Stroboscope and Mo- 
tion Pictures, W. A. Palmer, Camera Craft, 
45: 31-34, Jan. 1938. 

Twenty Years of Development of High- 
Frequency Cameras, H. E. A. Joachim, 
Jour. SMPE, 30: 169-80, Feb. 1938. 

Direct Photography of Ultracentrifuge Sedi- 
mentation Curves, J. St. L. Philpot, 
Nature, 141: 283-84, Feb. 12, 1938. 

Back-Projection for Amateur Studios, R. H, 
Alder, Amat. Cine World, 4: 607-8, 629. 
Mar. 1938. 

Seeing the Unseen, R. M. Horn, Mech. Eng., 
60: 199-201, Mar. 1938. 

Adapt Your Movie Camera for Speed Stills, 
S. A. Pearl, Pop. Phot., 2: 35, 87-88, Mar. 
1938. 

Photomicrography of Moving Objects by 
Electronic Photoflash Technique, S. H. 
Webster, E. J. Liljegren and D. J. Zimmer, 
J. Biol. Phot. Assoc., 16: 99-102, Mar. 
1938. 

Photographic Study of the Motion of Fibers 

and Water in Flowing Fiber Suspensions, 

L. A. Moss and E. O. Bryant, Paper Trade 

J., 106: 46-57, Apr. 14, 1938. 
Arc Studies with High-Speed Camera, 

Suits, J. Appl. Phys., 9: 188-89, Mar. 

1938; cf. Brit. J. Phot., 85: 362, June 10, 

1938. 
Studying Engine Combustion by Physical 

Methods, L. Withrow and G. M. Rass- 

weiler, J. Appl. Phys., 9: 362-72, June 

1938. 
Tension-Impact Testing, D. S. Clark and G. 

Datwyler, Mech. Eng., 60: 559, July 1938. 
It's a Photographic Finish, H. C. Grantz, 

Cinema Digest, 4: 8, 13, July 1938. 
Hochfrequenz-Kinematographie und Funk- 

enphotographie als Prufmethoden bei der 

Wassenfabrikation, Geissler, Filmtechnik, 

14: 206-7, Dec. 1938. 
The Photography of Airscrew Sound Waves, 

W. F. Hilton, Proc. Roy. Soc., 169A: 

174-90, #937, Dec. 22, 1938, (Modified 

form of spark photography.) 



High-Frequency Photography and Aircraft 
Design, Brit. J. Phot., 86: 809, Dec. 23, 
1938. (Film made by A. Lippisch pro- 
jected before Roy. Aeronautical Soc. showing 
air-flow forms. 1000 frames per sec.) 

A Study of Air Movement through Axial- 
Flow Free-Air Propellers, M. F. Dowell, 
Gen. Elec. Rev., J&: 210-17, 1939. 

Direkte photographische Aufnahme von 
Elektrophorese-Diagrammen, H. Svens- 
son, Roll. Zeit., 87: 181-86, #2, 1939. 

Overhead Welding, J. Sack, Philips Tech. 
Rev., 4: 9-15, Jan. 1939. 

Super-Super Speed, N. Taylor, Minicam, 

2: 58-59, Jan. 1939. (Golf pictures at 

1 1 100, 000th, 1 1 100th and 1 / '600th sec. 
intervals.) 

Elektronen-Vierstrahlroehre hoher Schreib- 
geschwindigkeit, A. Bigalke, Arch, fur 
Elektrotech., 33: 108-18, Feb. 15, 1939, 
(The maximum photographic recording speed 
is 1000 km /sec.) 

The Flashing Stroboscope, /. Sci. Instr., 
15: 398, Dec. 1938; Brit. J. Phot., 86: 252, 
Apr. 21, 1939. 

The Testing of Electric Fuses with the 
Cathode-Ray Oscillograph, J. A. M. 
Liempt and J. A. deVriend, Philips Tech. 
Rev., 4: 118-20, Apr. 1939. 

Electrophoresis of Proteins by the Tiselius 
Method, L. G. Longsworth and D. A. Mac- 
Innes, J. Phys. Chem., 24: 271-81, Apr. 
1939. 

High-Speed Cinematography in Post Office 
Engineering, R. W. Palmer, J. Brit. Kine- 
mat. Soc., 2: 119-24, Apr. 1939. (Solu- 
tion of problems in telephone engineering.} 

Photographische Flugleistungsmessungen, 
Phot. Ind., 37: 639-41, May 24, 1939. 

Physics of Flames and Explosions of Gases, 
B. Lewis and G. von Elbe, J. Appl. Phys., 
10: 344-59, June 1939. 

Glass Fracture Velocity, F. E. Barstow and 
H. E. Edgerton, J. Amer. Ceramic Soc. 
22: 302-7, Sept. 1939. 

Engine Flame Researches, T. A. Boyd, J. 
Soc. Automotive Engr., 45: 421-32, Oct. 
1939. 

Multiflash Photography, H. E. Edgerton, 
K. J. Germeshausen and H. E. Grier, 
Photo Technique, 1: 14-16, #5, Oct. 1939. 

Jets gazeux a vitesse supersonique et lumin- 
osit6s de detonation, H. Muraour, Chimie 
et Industrie, 42: 604-6, Oct. 1939. 

Super Stroboscope, Kern & Co., Ltd., J. 
Sci. Instr., 16: 381-82, Dec. 1939. 



Bibliography on High-Speed Photography 



105 



Photographische Untersuchungen an Flam- 
men, W. Jost, Zeit. angew. Phot., 2: 1-8, 
#1 /2, 1940. (Method for these high-spe.ed 
photographs .) 

Modern Synchroflash Photography, W. D. 
Morgan, Amer. Ann, Phot., 54: 203-18, 
1940. (Details on lamp and synchronizers.) 

Recording Transient Disturbances, O. D. 
Grismore, Bell Lab. Record, 18: 140-44, 
Jan. 1940. 

High-Speed Motion Pictures of the Human 
Vocal Cords, D. W. Farnsworth, Bell Lab. 
Record, 18: 203-8, Mar. 1940. 

Photographing Arcs Under Water, Product 
Eng., 11: 136-37, Mar. 1940. 

A Recording System Designed for the In- 
vestigation of the Electrical Relations in 
the Brains of Small Animals, G. A. Woon- 
ton. Can. J. Research, ISA: 65-73, Apr. 
1940. (Two R.C.A. high-vacuum cathode- 
ray tubes and a motor-driven Stoppani 
camera were used.) 

Taken at 1 /100.000 of a Second, A. King and 
G. Mili, Minicam, 3: 18-25, May 1940. 
(Edgerton ' ' speedlite . ") 

Methods of Measuring the Torque of In- 
duction Motors, A. C. W. V. Clarke, J. 
Inst. Elec. Eng. (London), 86: 597-600, 
June 1940. 

High-Speed Photography and the Study of 
Rapid Machine Motion, V. Sepavich and 
A. Palmer, Mech. Eng., 62: 519-24, July 
1940. 

High-Speed Photography Applied to the 
Study of Machine Performance, E. M. 
Watson, Product Eng., 11: 340-43, Aug. 
1940. 

Photoelasticity, an Application of Polarized 
Light to the Study of Strains and Stresses, 
M. Hetenyi, Photo Technique, 2: 46-49, 
Nov. 1940. 

Measurements on Detonation in the Duch6ne 
Apparatus, G. Broersma, /. Aeronautical 
Sci., 8: 62-72, Dec. 1940. 

Ueber kinematographische Aufnahmen von 
Pfeifentonen, K. T. Kuhn, Zeit. angew. 
Phot., 2: 92-96, #5/6, Dec. 1940. 

Cinematographic des mouvements ultra- 
rapides, M. L. Bull, Universite de Paris, 
Paris, 16 pp., 1941. 

Ballistics Photography, High Speed, R. E. 
Evans, Complete Phot., 1: 383-87, Issue 6, 
1941. 

Measurement of Supersonic Absorption in 
Gases by Optical Methods (in Russian), 
F. A. Korolev, J. Expt. & Theoret. Phys., 
11: 184-93, #1, 1941. 

Race-Track Photofinish Photography, R. F. 
Mitchell, Amer. Annual of Photography, 
66: 183-88,1941. 



Color Pictures of Photoelastic Models by 
Polarized Light, G. E. Holm, Photo Tech., 
3: 65-66, Jan. 1941. 

The Dynamics of Sneezing-Studies by High- 
Speed Photography, M. W. Jennison, 
Sci. Monthly, 62: 24-33, Jan. 1941. 

Photographing Electric Arcs in Liquids, T. E. 
Browne, Jr., Photo Tech., 3: 52-54, Feb. 
1941. 

Further Studies of Glass Fracture with High- 
Speed Photography, H. E. Edgerton and 
F. E. Barstow, J. Amer. Ceramic Soc., 
24: 131 -37, Apr. 1941. 

The Public Health Applications of High- 
Speed Photography, C. E. Turner, M. W. 
Jennison and H. E. Edgerton, Amer. J. 
Public Health, 31: 319-24, Apr. 1941. 

Stopping Motion for Design Data, K. D. 
Moslander, Machine Design, 13: 51-54, 
June 1941. 

Tools Seen in Slow Motion; G. E. Engineers 
Find High-Speed Photography Useful in 
Answering Questions on Chip Formation, 
L. T. Weller and E. M. Watson, Amer. 
Machinist, 85: 689-92, July 1941. 

Photographic Study of A. C. Arcs in Flowing 
Liquids, J. Slepian and T. E. Browne, Jr., 
Elec. Eng., 60: 823-38, Aug. 1941. 

Diesel Nozzle Spray : A Procedure for Deter- 
mining Discharge Characteristics, H. F. 
Bryan, Automobile Eng., 31: 337-40, 
Oct. 1941. 

Aids for Analyzing High-Speed Action, E. M. 
Watson, Gen. Elec. Rev., 44: 549-57, Oct. 
1941. 

The Direct Measurement of Lightning Cur- 
rent, J. W. Flowers, /. Franklin Inst., 232: 
425-50, Nov. 1941. 

High-Speed Photography Aids in Metal- 
Cutting Tests, Pop. Phot., 9: 87, Nov. 
1941. 

High-Speed Photography and Its Applica- 
tion to Industrial Problems, E. D. Eyles, 
J. Sci. Instr., 18: 175-84, Sept. 1941; 
244, Dec. 1941. Abridged in: J. Brit. 
Kinemat. Soc., 5: 114-27, Oct. 1942. 

Study of High-Speed Photography of Com- 
bustion and Knock in a Spark-Ignition, 

C. D. Miller, Report #727 of Nat. Advisory 
Committee for Aeronautics, 23 pp., 1942. 

Elektrophoretic and Ultracentrifugal Analy- 
sis of Hay-Fever-Producing Component of 
Ragweed Pollen Extract, H. A. Abramson, 

D. H. Moore and H. H. Gettner, J. Phys. 
Chem., 46: 192-203, Jan. 1942. 

Visual Studies of Flow Patterns, E. A. Hauser 
and D. R. Dewey, J. Phys. Chem., 46: 
212-13, Jan. 1942. 



106 



January 1951 Journal of the SMPTE Vol. 56 



The Air Corps' Newest Camera Gun, R. N. 

Haythorne, Amer. Cinemat., 23: 11, 37-38, 

Jan. 1942. 
Lip Vibrations in a Cornet Mouthpiece, 

D. W. Martin, J. Acoustical Soc. Amer., 
IS: 305-8, Jan. 1942. 

Balancing of Locomotive Reciprocating 
Parts, E. S. Cox, Engineering, 153: 36-38, 
Jan. 9, 1942. 

Shutters and Cutters. High-Speed Motion- 
Picture Photography Used in Study of 
Machine-Tool Cutting Edges, B. L. 
McKenzie, Sci. Amer., 166: 60-62, Feb. 
1942. 

A Method of Recording Low-Intensity 
Flashes of Light, C. Butt and R. S. Alex- 
ander, Rev. Sci. Instr., 13: 151-53, Apr. 
1942. 

"Shooting" for Keeps, C. J. Duncan, Mini- 
cam^: 64-67, May 1942. 

Photographers Play with Lightning, J. W. 
Gillon, J. Phot. Soc. Amer., 8: 217-23, 
May 1942. 

High-Speed Photographic Analysis of Metal 
Cutting, Machinery (London), 60: 521-22, 
June 4, 1942. 

The Production of Elastic Waves by Explo- 
sion Pressures, II., J. A. Sharpe, Geo- 
physics, 7: 311-21, July 1942. (A cathode- 
ray oscilloscope was used with ultraspeed 
film photosensitized with mercury.) 

Gun Camera Trains Pilots, Pop. Phot., 11: 
32, 72-73, Sept. 1942. 

Decay of Vibration Phenomena of Glass 
Bars, J. G. McCann, J. Amer. Ceramic 
Soc., 26: 409-13, Oct. 1, 1942. 

The Determination of an Unknown Fre- 
quency from a Photographic Record, M. 
Scott, Electronic Eng., 15: 243, Nov. 
1942. 

The Photographic Analysis of Motion, E. R. 
Davies, Proc. Roy. Inst. Great Britain, 32: 
384-98, Dec. 1942. 

Photographic Striation Methods Applied 
to the Supersonic Wind Tunnel of the 

E. T. H., Zurich, P. de Haller, E. T. H. 
Proceedings, 44-49, #8, 1943. 

Identification of Knock in NACA High- 
Speed Photographs of Combustion in a 
Spark-Ignition Engine, C. D. Miller and 
H. Lowell Olsen, National Advisory Com- 
mittee for Aeronautics. Report #761, 1943. 

Speedlamp, M. L. Sandell, Complete Phot., 
9: 3281-88, Issue 51, 1943. 

Applications of Photography in Shipbuilding 
and Engineering, F. J. Tritton, Trans. 
Inst. Engineers and Shipbuilders in Scot- 
land, 86: 135-64, #6, 1943. 



Visualizing High-Speed Action, E. M. Wat- 
son, Business Screen Mag., 4: 22-6, #8, 
1943. 

Recording Unit for Strain and Timing Func- 
tions, J. H. Meier, Electronics, 16: 79-83, 
114, Apr. 1943. 

Air Flow Visualization Opens New Avenues 
of Research, E. J. Saxl, Aviation, 42: 
148-51, 325, 329, Apr. 1943. 

Photographs of Sprays May Increase Gaso- 
line Mileage, Commercial Phot., 18: 246, 
Apr. 1943. (Camera used to determine 
size of drops entering engine of car.} 

Simple Time Base for a High-Speed Cine 
Camera, E. D. Eyles, J. Sci. Instr., 20. 
114-15, July 1943. 

Some Applications of High-Speed Photog- 
raphy, E. D. Eyles, Phot. J., 83: 261- 
65, July 1943. 

120,000 Frames per Second, F. Luther, 
Minicam Phot., 7: 36-47, Sept. 1943. 

An Inexpensive Stroboscope for High-Speed 
Photography, S. Silverman and W. H. 
Warhus, Rev. Sci. Instr., 14: 273-75, 
Sept. 1943. 

Effect of High-Intensity Arcs upon 35-mm 
Film Projection, E. K. Carver, R. H. Tal- 
bot and H. A. Loomis, Jour. SMPE, 41: 
69-87, July 1943; Int. Projectionist, 18: 
10-12, Sept. 1943; 12-14, 19, Oct. 1943. 

Four-Microsecond Flash Unit, Electronics, 
16: 144, 146, Oct. 1943. 

The Photographic Determination of Flame 
Temperatures in Closed- Vessel Explosions, 
A. S. Leah, Phil. Mag. (7), 38: 795-803, 
Dec. 1943. 

Technique of Stroboscopic Studies of Insect 
Flight, C. M. Williams, Science, 98: 
522-24, Dec. 1943. 

Examination of the Bursting Test by High- 
Speed Cinematography, H. A. Harrison 
and G. F. Underhay, Proc. Tech. Sect. 
Paper Makers' Assn., 25: 273-79, 1944. 

Study of High Explosives by High-Speed 
Photography, R. W. Cairns, Ind. Eng. 
Chem., 36: 79-85, Jan. 1944. 

Industrial Research Progress at the Armour 
Research Foundation 1942-43: High- 
Speed Photography, Chem. Eng. News, 22: 
29, Jan. 10, 1944. 

Camera Records High Speeds, A. L. Gale, 
Movie Makers, 19: 54, 76-78, Feb. 1944. 

High-Speed Movement, Aircraft Production, 
6: 125-28, Mar. 1944. 

Movies of Bullets, R. H. Bailey, Amer. Cine- 
mat., 25: 11, Apr. 1944. (Aerial gun 
camera.) 



Bibliography on High-Speed Photography 



107 



Cameras Synchronized with Guns Record 
Hits of Aerial Gunners, Phot. Trade News, 
8: 23-24, Apr. 1944. 

Rate of Rise of Water in Capillary Tubes, 
W. A. Rense, J. Appl. Phys., 15: 436-37, 
May 1944. 

On the Destructive Action of Cavitation, 
M. Kornfeld and L. Suvorov, /. Appl. 
Phys., IS: 495-506, June 1944. 

Photographic Study of the Retraction of 
Stressed Rubber, B. A. Mrowca, S. L. 
Dart and E. Guth, Phys. Rev., 66: 32-33, 
July 1 and 15, 1944. 

Persistence of Luminosity in Air, J. J. O'Do- 
herty, Nature, 154: 339, Sept. 9, 1944. (32 
frames per sec.) 

Aids for Pictorially Analyzing High-Speed 
Action, E. M. Watson, Jour. SMPE, 43: 
267-88, #4, Oct. 1944. 

Harnessing Time and Motion for Industrial 
Research, H. D. McLarty, Iron and Steel 
Eng., 21: 70-73, Dec. 1944. 

Possibilites industrielles d'un laboratorie 
de photographic et cinematographic ultra- 
rapides, P. Fayolle, Trav. et mem. du labora- 
torie central de* ind. mech., 17-25, #1, 1945. 

Retraction and Stress Propagation in Na- 
tural and Synthetic Gum and Tread 
Stocks, B. A. Mrowca, S. L. Dart and E. 
Guth, J. Appl. Phy8., 16: 8-19, Jan. 
1945. 

Photographing the Invisible, A. M. Lavish, 
Minicam Phot., 8: 22-24, #5, Feb. 1945. 

A Special Problem in Time-Microscopy, H. 
M. Lester, J. Phot. Soc. Amer., 11: 65-68, 
Feb. 1945. 

Samuel Slater Memorial Textile Research 
Laboratory, E. R. Schwarz, Textile Re- 
search J., 15: 33-36, Feb. 1945. 

Formation of Metal-Sprayed Deposits, W. E. 
Ballard, Proc. Phys. Soc. (London), 67: 
67-83, Mar. 1945. 

Photography as Applied to Textile Research, 
C. W. Bradley, J. Soc. Dyers and Colour- 
ists, 61: 61-64, Mar. 1945. (Electric 
spark used at Shirley Inst. Also mentions 
cathode-ray briefly.) 

How Does a Fly Land on the Ceiling, E. D. 
Eyles, Proc. Roy. Entomolog. Soc. London, 
20A: 14-15, Pts. 1-3, Mar. 1945. 

Triggering Speedlights, A. Palme, Amer. 
Phot., 39: 34-36, May 1945. 

How High-Speed Photography Aids in Re- 
designing, W. S. Calvert and H. D. Jackes, 
Machine Design, 16: 133-38, Feb. 17, 
1945; Brit. J. Phot., 92: 198, June 15, 
1945. 

High-Speed Linear Photography, E. N. 
Harvey and F. J. M. Sichel, J. Cett. 
Comp. Physiol., 26: 175-79, June 1945. 

Modern Measurement of Projectile Speeds, 



T. H. Johnson, Electronic Ind., 4: 82-85, 
170, 174, 178, 182, 186, 190, July 1945. 

High-Speed Cinematography, F. C. Johan- 
sen, J. Inst. Mech. Eng., 152: 224-25, 
Sept. 1945. 

The Recording of Strain by the "Parallel 
Resonance" Method, H. J. Beach, Elec- 
tronic Eng., 17: 737, Oct. 1945. 

Aerodynamic Performance of Small Spheres 
for Subsonic to High Supersonic Velocities, 
A. C. Charters and R. N. Thomas, J. 
Aeronautical Sci., 12: 468-76, Oct. 1945. 

Cinematographic Indicator for Locomotive 
Engine Cylinders, G. Bohl, Engineers' 
Digest, 2: 602-4, Dec. 1945 (taken from 
Rev. gen. des chemin de fer, 63: 137-44 
Nov.-Dec. 1944) . 

Slow-Motion Study of Injection and Com- 
bustion of Fuel in a Diesel Engine, C. D. 
Miller, S. A. E. Journal, 53: 719-35, 
Dec. 1945. 

Photoballistic Method of Studying Very 
Short Bursts of Light (In Russian), V. 

A. Tsukerman, Bull. Acad. Sci. (USSR), 
Dept. Sci. Tech., #6: 863-74, 1946. 

A Combined Method for Photographing Very 
Rapid Processes (In English), V. A. Tsuk- 
erman, Compt. rend. (Doklady), acad. sci. 
(USSR), 53: 319-21, #4, 1946. 

Relation between Spark-Ignition Engine 
Knock Detonation Waves and Autoigni- 
tion as Shown by High-Speed Photography, 
C. D. Miller, National Advisory Committee 
for Aeronautics Wartime Report #E-238, 
1946. (P.B. 96,955 issued in 1949.) 

Apparatus for Stroboscopic Observations, 
S. L. deBruin, Philips Tech. Rev., 8: 25-32, 
Jan. 1946. (A stroboscope of an argon gas 
discharge lamp and electric apparatus for 
condenser discharges.) 

How a Bullet Shatters Glass, A. Palme, 
Amer. Phot., JO: 14-15, Jan. 1946. 

Electronic Timing of Sequence Photographs. 
C. H. Coles, Electronic Ind., 6: 74-76, 
Feb. 1946. 

Materials in Action as Seen by the High- 
Speed Camera, K. Rose, Materials and 
Methods, 23: 414-17, Feb. 1946. 

Single-Blow Impact Test for Cast Iron, A. 

B. Everest, J. W. Grant and H. Morrogh, 
Engineering, 161: 117-20, Feb. 1, 1946. 

Specialized Photography Applied to Engi- 
neering in the Army Air Forces, P. M. 
Thomas and C. H. Coles, Jour. SMPE, 
46: 220-30, Mar. 1946. (High-speed mo- 
tion pictures, high-speed still pictures, spe- 
cial recording devices.) 

High-Speed Photographs of Under-Water 
Explosions, D. S. Senior, Phot. J., 86B: 
25-31, Jan.-Feb. 1946; cf. Chem. Prod- 
ucts, 9: 52-58, Mar.-Apr. 1946 and Elec- 
tronic Eng., 18: 133-35, May 1946. 



108 



January 1951 Journal of the SMPTE Vol. 56 



Initial Stages of the Explosion of Nitrogly- 
cerine, R. G. Vines and M. F. R. Mul- 
cahy, Nature, 157: 626, May 11, 1946. 

Photographic Uses of Electrical Discharge 
Flash-Tubes, H. E. Edgerton, J. Opt. Soc. 
Amer., 86: 390-99, July 1946. 

"And Here Are the Pictures That Show 
It!" (Rifle Recoil after Shot Is Fired), 
C. H. Hopkins, Phot. Age, 1: 16-17, #1, 
July 1946. 

You Can Advance Your Industrial Research 
and Production with High-Speed Motion 
Pictures, Phot. Age. 1: 34-37, 58, July 
1946. 

Instrument for Investigating the Operation 
of Camera Shutters, D. R. Dighton and 
H. M. Ross, Phot. J., 86B: 110-16, Sept- 
Oct. 1946. 

Fastax at Bikini, J. H. Waddell, Bell Lab- 
Record, 24: 358-62, Oct. 1946. 

Greatest Photographic Organization in His- 
tory. Shot Bikini Blast, G. Warrenton, 
Amer. Cinemat., 27: 352, 383, Oct. 1946. 

Measuring Temperatures by Photography, 
J. H. Hall, Phot. Trade Bull., 7: 619, Oct. 
1946. (Exposures of 1.8 milliseconds re- 
vealed formations in steel.) 

How the Atom-Bomb Tests Were Photo- 
graphed (Operation Crossroads), H. L. 
Hansen and D. P. Schiedt, U. S. Camera, 
9: 16-19, 42, 48, Oct. 1946. 

Action in a Split Second, U. S. Camera, 9: 
15, Nov. 1946. (Hockey games. Lighting 
mentioned?) 

Roles of Detonation Waves and Autoigni- 
tion in Spark Ignition Engine Knock as 
Shown by Photographs Taken at 40,000 
and 200,000 Frames per Second, C. D. 
Miller, Trans. Soc. Automotive Eng., 1: 
98-143, Jan. 1947; abridged in J. Soc. 
Automotive Eng., 54: 34-35, Oct. 1946. 

Photographic Analysis of Performance, G. A. 
Jones, Phot. J., 87B: 7-12, Jan.-Feb. 
1947. 

Proceedings of the Thirteenth Annual Meet- 
ing (of the Amer. Soc. of Photogram- 
metry: Photography at Operation Cross- 
roads), P. T. CuUen, Photogrammetric 
Eng., 13: 95-100, Mar. 1947. 

Spread around the Initiating Point of the 
Detonation Wave in High Explosives, 
Nature, 159: 402, Mar. 22, 1947. 

Cathode-Ray Tube Shutter-Testing Instru- 
ment, D. T. R. Dighton and H. McG. 
Ross, /. Sci. Instr., 24: 128-33, May 1947. 

The Shadowgraph Recording Method, D. L. 
Supernaw, Instruments, 20: 532-35, June 
1947. 

Special Cameras and Flash Lamps for High- 
Speed Underwater Photography, R. T. 
Knapp, Jour. SMPE, 49: 64-82, July 
1947. 



Photography in Engine Research, H. D. 
Goulding, Aircraft Production, 9: 283-87, 
Aug. 1947; 338-42, Sept. 1947. 

High-Speed Photography. Pt. I. External 
Surfaces and Opaque Objects, G. A. 
Hawkins and C. E. Balleisen, Machine 
Design, 19: 127-33, Aug. 1947; Pt. II. 
Radiography and Motion Pictures, Ma- 
chine Design, 19: 121-26, Sept. 1947. 

Measuring Pressures of Industrial Explo- 
sions, N. J. Thompson and E. W. Clusins, 
Electronics, 20: 90-93, Nov. 1947. 

Velocity of Detonation of a Tubular Charge 
of Explosive, D. W. Woodhead, Nature, 
160: 644, Nov. 8, 1947. 

High-Speed Movie Solves Tricky Problem in 
Perfecting High-Pressure Hydraulic Trans- 
mission, S. G. Guins, Machine Design, 19: 
126-28, Dec. 1947. 

Underwater Explosions, R. H. Cole, Prince- 
ton University Press, Princeton, pp. 210- 
28, 1948. (On photography of underwater 
explosions.) 

The Use of High-Speed Motion Pictures in 
Ordnance Research, J. E. Brock, abridg- 
ment in: Proc. Indiana Acad. Sci., 57: 
183-85, 1948. (Uses Fastax Camera.) 

Kinematographie und vollautomatische Se- 
rienaufnahmen schnellveranderlicher Elek- 
troneninterferenzen, G. Mollenstedt, Op- 
tik, 8: 68-74, #12, 1948. 

Film Solves Mystery as to How a Fly Flies, 
Film World, 4: 39, 54, Jan. 1948. 

Photographic Techniques of Combustion 
Research, L. Beral and W. T. Cooper, 
Fuel, 27: 10-18, Jan.-Apr. 1948. 

A Cinema-Spectrograph for Photographing 
Rapid Spectral Sequences, R. C. Herman 
and S. Silverman, J. Opt. Soc. Amer., 38: 
209-11, Feb. 1948. 

High-Speed Motion Pictures with Synchro- 
nized Multiflash Lighting, R. A. Anderson 
and W. T. Whelan, Jour. SMPE, 50: 109- 
208, Mar. 1948. 

Looking at Flash, E. L. Auld, Pop. Phot., 
22: 58-59, 164-68, Mar. 1948. 

Electronic Stroboscopes, A. E. Crawford, 
Aircraft Production, 10: 101-2, Mar. 1948. 
(Basic operating principles of short dur- 
ation flash equipment as alternative to high- 
speed photography.) 

Circus Action in Color, H. E. Edgerton, 
National Geographic Mag., 48: 305-8 and 
plates, Mar. 1948. 

Electronic Flashtube Illumination for Spe- 
cialized and Motion Picture Photography, 
H. M. Lester, Jour. SMPE, 50: 208-32, 
Mar. 1948. 

Microflash Unit for Ballistic Photography, 
W. W. McCormick, L. Madansky and 
A. F. Fairbanks, J. Appl. Phys., 19: 221- 
25, Mar. 1948. 



Bibliography .on High-Speed Photography 



109 



Photography in the Study of Fibers and 
Textiles, C. W. Bradley, Fibres, 9: 45-49, 
95-98, 138-41, Feb., Mar., Apr. 1948. 

High-Speed Photography of Welding Arcs, 
F. Brailsford and K. F. Shrubb, J. Sci. 
Instr., 25: 211-13, June 1948. 

Motion Picture Photography of Television 
Images, R. Fraser, RCA Rev., 9: 202-17, 
June 1948. (Used to record television 
transmissions from cameras in aircraft and 
guided missiles.) 

Knocking Combustion Observed in a Spark- 
Ignition Engine with Direct and Schlieren 
High-Speed Motion Pictures and Pressure 
Records, G. E. Osterstrom, National Ad- 
visory Committee for Aeronautics, Tech. 
note #1614, June 1948. 

Photographic Tracking of Guided Missiles, 
L. M. Biberman, S. E. Dorsey and D. L. 
Ewing, Electronics, 21: 92-95, July 1948. 

Tracking Rockets, M. Mann, Ordnance, 32: 
23-25, July-Aug. 1948. 

High Speed Photographs of Water Jet in 
the End-Quench Test, R. A. Buchanan, 
Metal Progress, 64: 180-81, Aug. 1948. 

Aids to Analysis of Patterns Obtained in the 
Diffraction of Electrons by Gases, H. J. 
Yearian and W. M. Barss, J. Appl. Phys., 
19: 700-4, Aug. 1948. 

Measurement of the Temperature of Sliding 
Surfaces with Particular Reference to Rail- 
way Brake Blocks, R. C. Parker and P. R. 
Marshall, Proc.Inst. Mech. Eng., 168: 209- 
29, Sept. 1948. 

Highlights and Side Lights: Photographs at 
High Speed, Gen. Elec. Rev., 51: 47-48, 
Sept. 1948. 

How the Camera Solves Tough Plant Prob- 
lems, Modern Industry, 16: 50-53, Sept- 
1948. (General discussion on uses, lamps, 
etc.) 

Full Scale Free-Flight Ballistic Measure- 
ments of Guided Missiles, L. A. Delasso, 
L. G. DeBey and D. Reuyl, J. Aeronautical 
Sci., 15: 606-15, Oct. 1948. 

High-Speed Photography, A Useful Tool, E- 
P. Wightman, PSA Jour., 14: 668, Nov. 
1948. 

Underwater Photographs of Flow Patterns 
for Surface Vessels, W. H. Sutherland, 
Marine Engineering, 68: 64-67, Oct. 1948; 
cf. Mech. Eng., 70: 1004-5, Dec. 1948. 

The Flash-tube and Its Applications, J. N. 
Aldington and A. J. Meadowcroft, J. Inst. 
Elec. Eng., 95: 671-81, Pt. 2, Dec. 1948. 

A Working Manual for Spark Shadowgraph 
Photography, B. S. Melton, R. Prescott 
and E. L. Gayhart, Bumblebee Series, 
Report #90, Johns Hopkins University, 
Applied Physics Laboratory, Dec. 1948. 



Electronic Flash (Gas Discharge) Tube in 
Photography of the Anterior Segment of 
the Eye, R. R. Trotter and W. M. Grant, 
Arch. Ophthalmology, 40: 493-96, Nov. 
1948. 

The Film in Scientific Research, G. A. Jones, 
British Sci. News, 2: 235-38, #20, 1949. 
(Photoelastic patterns in a railway line and 
chair; cylinder of combustion engine.) 

Photographs at 500,000 frames per Second of 
Combustion and Detonation in. a Re- 
ciprocating Engine, T. Male, pp. 721-26, 
in: Third Symposium on Combustion and 
Flame and Explosion Phenomena, Williams 
& Wilkins, Baltimore, Md., 1949. 

Technique for the Optical Measurement of 
Turbulence in High-Speed Flow, L. S. G. 
Kovasznay, pp. 211-222, in: Heat Trans- 
fer and Fluid Mechanics Institute, 1949. 
American Society of Mechanical Engineers, 
N.Y. 

Use of High-Speed Photography in the Air 
Forces, E. A. Andres, Sr., Jour. SMPE, 
52: 81-89, Mar. Pt. 2, 1949. 

Applications of High-Speed Photography, 
M. Beard, Jour. SMPE, 52: 97-106, 
Mar. Pt. 2, 1949. 

Control Unit for Operation of High-Speed 
Cameras, L. L. Neidenberg, Jour. SMPE, 
52: 107-9, Mar. Pt. 2, 1949. 

Methods of Analyzing High-Speed Photo- 
graphs, W. S. Nivison, Jour. SMPE, 
52: 49-60, Mar. Pt. 2, 1949. 

High-Speed Photography in the Automotive 
Industry, R. O. Painter, Jour. SMPE, 52: 
90-96, Mar. Pt. 2, 1949. 

High-Speed Photographic System Using 
Electronic Flash Lighting, W. T. Whelan, 
Jour. SMPE, 52: 116-29, Mar. Pt. 2, 
1949. 

Electrical-Flash Photography, H. E. Edger- 
ton, Jour. SMPE, 52: 8-23, Mar. Pt. 2, 
1949. 

High-Speed and Time-Lapse Photography 
in Industry and Research, H. M. Lester, 
Jour. SMPE, 62: 71-80, Mar. Pt. 2, 1949. 

Splashes from Underwater Explosions, H. 
Kolsky, J. P. Lewis, M. T. Sampson, A. 
C. Shearman and C. I. Snow, Proc. Roy. 
Soc., 196A: 379^01, Apr. 7, 1949. 

A Visual Method for Demonstrating the Path 
of Ultrasonic Waves through Thin Plates 
of Material, C. J. Burton and R. B. Barnes, 
/. Appl. Phys., 20: 462-67, May 1949. 
(Photographs show reflection and trans- 
mission by metal plates.) 

Johns-Manville Research Center Uses High- 
Speed Photography as a Scientific Tool, 
Ted Czarda, Phot. Age, 4: 15-16, May 
1949. 



110 



January 1951 Journal of the SMPTE Vol. 56 



Film Recording of Television Transmissions 
for the Purpose of International Pro- 
gram Exchanges (in French), Y. L. Del- 
bord, Ann. Telecommun., 4- 190-201, 
June 1949. 

New Photographic Technique for Observ- 
ing Velocity Fields in Water Induced by 
the Entry of Solid Missiles, G. Birkhoff 
and T. E. Caywood, J. Appl. Phys., 20: 
646-59, July 1949. 

Analyzing Equipment Operation by Photo- 
graphic Measurement Techniques, S. M. 
Keen, Product Eng., 20: 119-22, July 1949. 

The Recording of Optical Transients, H. A. 
Prime, Proc. Inst. Elec. Eng., 96: Pt. 
II: 662-70, Aug. 1949. 

The Dynamics of Cavitation Bubbles, M. 
S. Plesset, J. Appl. Mech., 16: 277-82, 
Sept. 1949. 

Auroral Radiation in the 3,000-Megacycle 
Region, P. A. Forsythe, W. Petrie and 
B. W. Currie, Nature, 164: 453, Sept. 10, 
1949. 

The Propagation of Flame. Studies of 
Shadow Cones and Bubble Effects, J. W. 
Linnett, Chem. Age, 61: 345, Sept. 10, 
1949. (Report of a lecture before the British 
Association.) 

Photographic Techniques in Combustion 
Research, L. Beral, Phot. J., 89B: 98- 
107, Sept.-Oct. 1949. 

Repeating Flash, J. N. Aldington, Functional 
Phot., 1: 14-17, Oct. 1949. (Used for 
flight of bullets and birds.) 

Shock Waves, Otto Laporte, Sci. Amer., 181: 
14-18, #5, Nov. 1949. 

Recent British Equipment and Technique 
for High-Speed Cinematography, G. A. 
Jones and E. D. Eyles, Jour. SMPE, 53: 
502-14, Nov. 1949. 



Exposure Meter for High-Speed Photog- 
raphy, E. T. Higgons, Jour. SMPE, 53: 
545-48, Nov. 1949. 

Measuring Shock with High-Speed Motion 

Pictures, J. T. Muller, Jour. SMPE, 53: 

579-87, Nov. 1949. 
High-Speed Motion Pictures in Full Color, 

F. M. Tylee, Jour. SMPE, 53: 588-93, 

Nov. 1949. 

Light-Meter Uses with Electronic Flash, H. 
E. Edgerton, PSA Jour. Pt. 2, 1: 6-10, 
Jan. 1950. (Photographic Science and 
Technique.) 

Flash Photolysis and Spectroscopy. A New 
Method for the Study of Free Radical 
Reactions, G. Porter, Proc. Roy. Soc., 
200A: 284-300, Jan. 6, 1950. 

Underwater Photography, J. B. Collins, 
Phot. J., BOB: 24-31, Jan.-Feb. 1950. 

Measurement of Normal Burning Velocities 
of Propane-Air Flames from Shadow 
Photographs, J. W. Andersen and R. S. 
Fein, J. Chem. Phys., 18: 441-43, Apr. 
1950. 

High-Speed Motion Pictures in Textile Re- 
search, E. K. Fischer and J. C. Burnett, 
Textile Research J., 20: 259-69, Apr. 1950. 

Motor Racing Photography, G. C. Monk- 
house, Photographic /., 90A: 149-155, 
May 1950. 

Streak Photography, I. Vigness and R. C. 
Nowak, J. Appl. Phys., 21: 445-48, 
May 1950. 

The Infra-Red Photographic Evaluator, 
S. Horsley, Amer. Cinemat., 31: 196, 211- 
13, June 1950. 

The Pressurized Ballistics Range at the Naval 
Ordnance Laboratory, L. P. Gieseler, Jour. 
SMPTE, 55: 53-59, July 1950. 



vii. X-RAY 



X-Ray Photographs with Extremely Short 
Exposure Times, W. J. Oosterkamp, 
Philips Tech. Rev., 5: 22-25, Jan. 1940. 

Ultra-Speed Radiography, L. F. Ehrke and 
C. M. Slack, Photo Technique, 3: 53-55 
Jan. 1941. 

Medical Physics, O. Glasser, ed., Year Book 
Publishers, Inc., Chicago, 1944; Radi- 
ography at High Speed by C. M. Slack, 
pp. 1185-87. 

One-Millionth-Second Radiography, C. M. 
Slack, J. Phot. Soc. Amer., 11: 302-6, 
Sept. 1945. 

Methods of Synchronization for Very Brief 
X-Ray Exposures (In Russian), G. M. 
Strakhovskii and V. A. Tsukerman, Bull. 
Acad. Sci. (USSR), Dept. Sci. Tech. *3: 
371-84, 1946. 



One-Millionth-Second Radiography and Its 

Applications, C. M. Slack and D. C. 

Dickson, Proc. Inst. Radio Eng., 35: 600-6, 

June 1947. 
A New Apparatus for Direct Cineroent- 

genography, A. S. Gidlund, Ada Radio- 

logica, 32: 81-88, #2/3, 1949. 
New Developments in X-ray Motion Pic- 
tures, C. M. Slack, L. F. Ehrke, C. T., 

Zavales and D. C. Dickson, Jour. SMPE 

52: 61-70, Mar. Pt. 2, 1949. 
Flash Radiography Applied to Ordnance 

Problems, J. C. Clark, J. Appl. Phys., 20: 

363-70, Apr. 1949. 
High-Speed Cine-Radiography, C. M. Slack, 

L. F. Ehrke, D. C. Dickson and C. T. 

Zavales, Non-Destructive Testing, 7: 7-11, 

23, Spring 1949. 



Bibliography on High-Speed Photography 



111 



A and B Windings of 16-Mm Raw-Stock Film 
With Perforations Along One Edge 



THE PROPOSED American Standard for 
Winding of 16-Mm Sound Film, was 
published as a first draft in the Septem- 
ber, 1949, JOURNAL. While that was the 
first time that winding 16-mm sound 
film had been proposed for adoption as 
an American Standard, the proposals 
were practices already followed by the 
film manufacturers for a number of 
years. It should also be noted that, in 
1941, the Society recognized the method 
of designating the two types of windings 
by publishing a Society recommendation, 
and that recommendation was substan- 
tially repeated hi the first draft for this 
proposed standard. 

As a result of publishing the first 
draft, comments were received which 



indicated ambiguity in the original 
wording; therefore, a second draft was 
prepared by the 16-Mm and 8-Mm 
Committee in January, 1950. That 
draft was sent to ballot of the 16-Mm 
and 8-Mm Committee in April, 1950. 
Only minor editorial comments were 
received and, therefore, this proposal 
is again being published for 90-day trial 
and comment (see page opposite). 

It is believed this standard fills a 
recognized need for uniform ways of 
designating the direction of winding of 
16-mm sound film. It is definitely not 
the intent of this standard to indicate 
any preference in the direction of wind- 
ing, since existing equipments are de- 
signed to use both styles. 



Revised American Standard Z22.40-1950 

Sound Records and Scanning Area 

of 35-Mm Sound Motion Picture Prints 



THIS STANDARD originated as an Ameri- 
can War Standard, Z52.36-1945. It 
was reapproved as American Standard, 
Z22.40-1946, in March, 1946 and pub- 
lished in the April, 1946 JOURNAL. How- 
ever, hi the republishing process, a 
minor drafting error occurred. The 



arrow pointing to the outer edge of the 
printed area fell slightly short of the 
outer edge. This revised standard, 
Z22.40-1950, corrects that error and is 
thus being published now as originally 
intended (see p. 114). 



112 



January 1951 Journal of the SMPTE Vol. 56 



Proposed American Standard 

AandBWindingsof16-Mm Raw-Stock Film 
With Perforations Along One Edge 



(Second Draft) 



Z22.75 



The purpose of this standard is to insure a 
uniform method of designating the types of 
winding (location of the perforated edge) in 
current use for 16-mm raw-stock film having 



perforations along one edge, thus to facilitate 
ordering and describing the film. 

With both types of winding described be- 
low, the emulsion side of the film shall face 
the center of the roll. 





Winding A 
Emulsion side in 



Winding B 
Emulsion side in 



When a roll of 16-mm raw stock perforated 
along one edge is held so that the outside end 
of the film leaves the roll at the top and 
toward the right, winding A shall have the 
perforations along the edge of the film toward 
the observer, and winding B shall have the 
perforations along the edge away from the 



observer. In either case, if the film is wound on 
a spool with a square hole in one flange and a 
round hole in the other flange, the square hole 
shall be on the side away from the observer. 

No preference for either type of winding is 
implied since both types are required for use 
on existing equipment. 



January 1951 Journal of the SMPTE Vol. 56 



NOT APPROVED 

113 



American Standard 

Dimensions and Locations for 
Sound Records and Scanning Area 

of 35-Millimeter Sound Motion Picture Prints 



Rr f . V. S. Pat. Og. 

Z22.40-1950 



vi.ion of 

232.40-194* 

tUDC 778.534.4 



AREA PRINTED 
IN SOUND PRINTER 



OPAQUE ON PRINT 



VARIABLE AREA 

AND MATTED 

VARIABLE DENSITY 

RECORDS 



OPAQUE ON PRINT 



FULL WIDTH 
VARIABLE DENSITY 
RECORD 



AREA SCANNED BY 
REPRODUCER 



a 

a 





n 


- j 


1 


i_r 



*-GUIDED EDG 

I 92 001 | N 


OUTER EDGE OF 


PRINTED AREA 
INNER EDGE OF 


1 


a 
o 
a 
n 


468 + 0.02 MM 
'< 0305i 002 IN 


PRINTED AREA 


775 005 MM 




r 



FECTIVE.L.Y " "J^f 




0.076000l IN. 


^y~ 
o 
o 



o 
jn 


: f\ 


r 




WIDTH OF 
SOUND RECORD \ 


MATTED V.D. \ 
(FULL WIDTH) AND \ 
IOO%MODULATIONVA d 

^ \ 


n ^* 

in 


1.931 0.02 MM 
0.2432|N. 




: : i 


t SOUND RECORD -4- 
"FECTIVELY ; 




6.17 0.05 MM 

o ioo!:So8 IN 




a i 


O 












WIDTH OF 


SOUND RECORD* 






in 


2 - 54 !?io MM 




: ____: 



SCANNED AREA 
WIDTH OF 



6.I7+0.02MM 
0.084 OOOI IN 



SCANNED AREA 
HEIGHT OF 



2.13 + 0.02 MM 



SCANNED AREA 



Distance Between Sound and Corresponding Picture The sound shall pre- 
cede the center of the corresponding picture by a distance of 20 Vz frames. 

These Dimensions and Locations Are Shown Relative to Unshrunk Raw Stock. 

'The only change in this standard over the 1946 edition is the correct posi- 
tioning of the arrows on the dimensions marked *. 



Approved October 6, 1950, by the American Standards Association, Incorporated 

Sponsor: Society of Motion Picture and Television Engineers tUni.i t>-,, m .i u....h, .,.,, 

Copyright, 1950, by American Standards Association, Inc.; reprinted by permission of the. copyright holder. 
114 January 1951 Journal of the SMPTE Vol. 56 



Edge Numbering 16-Mm Motion Picture Film 



THE PROPOSED American Standard for 
Edge Numbering 16-Mm Motion Pic- 
ture Film has been under discussion for 
several years. In the original discus- 
sions, neither the 16-frame nor the 40- 
frame interval could be unanimously 
chosen for a standard . The laboratories 
producing 16-mm prints from original 
35-mm material were desirous of retain- 
ing the 16-frame interval, while those 



working from original 16-mm film pre- 
ferred 40-frame separation of the 
numerals. In the latter part of 1949, 
there was further discussion in an at- 
tempt to establish the 40-frame interval 
as a standard. Consequently, this pro- 
posed standard has been written to 
make 16-mm edge numbering optional, 
but specifying that the 40-frame interval 
is preferred. 



Proposed American Standard 

Edge Numbering 16-Mm 
Motion Picture Film 



Z22.83 



The purpose of this standard is to establish 
a uniform practice with respect to the interval 
between edge numbers when they are latent- 
image printed on 16-mm raw-stock film. It is 
not intended to imply that all 16-mm film 
should be edge-numbered. 



The distance between consecutive numbers 
shall be 40 frames. Thus, the numbers will indi- 
cate film footage, subject to a small correction 
for shrinkage of the film. 



NOT APPROVED 



January 1951 Journal of the SMPTE Vol. 56 



115 



Tentative Recommendations 
for 16-Mm Review Rooms 
and Reproducing Equipment 



Foreword 



rriHE TENTATIVE RECOMMENDATIONS 

JL included herewith are the result of 
extensive work carried on by a Sub- 
committee of the 16- and 8-Mm Com- 
mittee under the Chairmanship of E. 
W. D'Arcy. It should be clearly under- 
stood that this is not a final Society 
recommendation, but rather that it is an 
interim report of the committee. The 
proposal is being published at this stage 
to make the information available to 
those who have use for it and to invite 
additional comments and discussion. 

At the outset it was agreed that the 
primary objective of the subcommittee 
was to establish a primary listening 
standard for gaging 16-mm print qual- 
ity. This decision was reached only 
after lengthy discussion regarding the 
possibility of actually specifying the 
ideal over-all response characteristic of 
16-mm portable projection equipment. 
To accomplish this end, however, ap- 
peared to be an insurmountable problem 
because of variation in response of the 
portable-type loudspeakers and the 
varying conditions under which the 
equipment is used. 

The problem, therefore, was ap- 
proached from the other direction, 
namely of trying to improve and make 



Third draft, dated November 15, 1950, 
edited and presented for publication on 
January 2, 1951. 



more uniform 16-mm release prints. 
It then would be left to the 16-mm pro- 
jector manufacturers to adjust their 
equipment in any way they saw fit to 
best reproduce these prints. 

In reaching this objective, listening 
tests were conducted employing various 
'wide-range two-way speaker systems 
as well as portable speakers normally 
used with 16-mm projectors. For those 
listening tests a wide selection of 16- 
mm release material was reproduced on 
these systems using a number of sug- 
gested frequency characteristics. 

As the testing progressed, it became 
more and more evident that : 

First, modification of the reproducer 
frequency characteristics from those 
recommended for 35-mm theater use by 
the Motion Picture Research Council 
produced little if any significant im- 
provement in the reproduction of 16-mm 
prints. 

Second, if a 16-mm print reproduced 
well, employing the 35-mm theater 
systems, it also reproduced well on a 
conventional 16-mm projector employ- 
ing small portable-type loudspeakers. 

Therefore, rather than try and es- 
tablish any particular electrical char- 
acteristic and portable-type speaker as 
a recommended review room system, 
it was agreed to accept those char- 
acteristics recommended for theater 
use, since many of the present 16-mm 



116 



January 1951 Journal of the SMPTE Vol. 56 



producing companies and processing 
laboratories already have 35-mm sys- 
tems in their review rooms which can 
readily be modified for the 16-mm re- 
production. 

The frequency-response characteris- 
tics shown on the following pages are 
identical with those established by the 
Motion Picture Research Council for 



use in reproducing 35-mm sound films in 
motion picture theaters. 

There are theater-type speakers not 
covered in these recommendations. 
It is hoped that suitable arrangements 
can be made for adding curves for these 
and future speakers in order that the 
recommendations may be as up-to-date 
and useful as possible. 



1 . Scope 



1.1: The purpose of this standard is 
to facilitate the production of 16-mm 
films having sound tracks of high, 
uniform quality. It is believed that the 
best way to attain this objective is to 
establish a reference system for judging 
the quality of the sound from 16-mm 
films . Such a system requires : 

(a) a sound reproducer having stand- 
ardized over-all electrical frequency- 
response characteristics, 

(b) a projector of high quality, and 

(c) a review room having good acous- 
tical properties. 1 



The characteristics established by 
the Motion Picture Research Council 2 
for various speaker systems used in re- 
producing 35-mm sound film have been 
adopted because extensive listening tests 
proved them to be optimum for repro- 
ducing sound from 16-mm film also. 

Thus, the electrical characteristic is 
specified for each particular speaker 
system. Before a system other than 
those shown below is used, it will be 
necessary to make comparative listening 
tests to determine the proper frequency- 
response characteristic for that system. 



2. Reproducer Requirements 



2.1: Power Output: The minimum 
power output of the reproducer amplifier 
shall be 15 w. The reproducer gain 
control should be calibrated in db and 
should indicate the gain setting re- 
quired to produce 10 w output from the 
American Standard Signal Level Test 
Film Z22.45. The amplifier should 
have enough available gain to produce at 
least 20 db in excess of that required to 
produce 10 w. 

2.2: Harmonic Distortion: The re- 
producer amplifier shall not introduce 
more than 1% harmonic distortion at 10 
w output and not more than 2% at 15 w 
at any frequency between 50 and 7000 
cycles. 



2.3: Signal-to-Noise Level: The over- 
all system noise measured electrically 
at the speaker terminals of the amplifier 
shall not be greater than 50 db. This 
measurement should be made with the 
system frequency response adjusted for 
the particular speaker in use and with 
the gain control set to deliver 10 w out- 
put when the American Standard "400 
Cycle Signal Level Test Film Z22.45" 
is passed through the projector. 

The output should be terminated in a 
noninductive resistive load equal to the 
nominal input impedance of the particu- 
lar speaker system in use. The meas- 
urement should then be made without 
film in the gate, with the projector 
running and exciter lamp turned on. 



"Theater acoustic recommendations of 
the Academy Research Council Theater 
Standardization Committee," Jour. SM- 
PE, vol. 36, Mar. 1941. 



2 "Standard electrical characteristics for 
theater sound systems," Motion Picture 
Research Council Bulletin, 1948 Volume. 



Recommendations for 16-Mm Review Rooms 



117 



24: Frequency Response: The over- 
all frequency-response characteristic of 
the projector and amplifier shall be ad- 
justed as shown hi Figs. 1 through 7, 
depending upon the theater speaker 
system to be employed. Variations are 
permitted from these nominal responses 
of =*= 1 db from 100 to 3000 cycles and 
increasing progressively with frequency 
to 2 db at 7000 cycles. To adjust the 
system to the reverberation character- 



istics of a specific review room, it may 
also be necessary to adjust the response 
below 100 cycles. Variation as great 
as 2 db are permitted as shown on the 
respective curves. 

These measurements shall be made 
with amplifier output terminated in a 
noninductive resistive load equal to the 
nominal input impedance of the par- 
ticular speaker system in use. The 
source of signal shall be a multifre- 



-# 






















































FREO 


DB 


1 






















































100 


3TO-I 




1 


m 


m 


f 































*^^ 


N^ 




^ 




^ 


V 






j 1000 
2000 
?500 

J500 
4000 




4 
io 

J 












1 


' 






































\ 




6000 
7000 
8000 


7 * 

-18 



200 JOO 400500 700 1000 
FREQUENCY IN CYCLES PER SECOND 



2000 JOOO 40006000 7000 10000 



Fig. 1. Recommended electrical characteristics for 16-mm review room re- 
producers employing Altec Lansing Energized Loudspeaker Systems Models 
75W5 and 30W5. High-frequency unit attenuation to 3 db. 



15 

10 





























































DB 

OTO-2 
OTO-I 


IN DECIBELS 
o o> 




s 











^Hmmmm 


MMMi 








Hi 





^^ 


" 










^^ 


B 


S 


. 


^s 


* 


s^- 










~ 100 
300 
. 500 
' 1000 
_ 2000 
: 2500 












I 














































4 


1 






- J500 
- 5000 


ij 

2 

-4 


14 




1 


















































v 

^ 




- 6000 


'i 

-13 
17 



200 JOO 400 500 700 000 
FREQUENCY IN CYCLES PER SECOND 



2000 JOOO 40005000 7000 IOOOO 



Fig. 2. Recommended electrical characteristics for 16-mm review room 
reproducers employing Altec Lansing Voice of the Theater Loudspeaker Sys- 
tems Models A1X, Al, A2X, A2, A4X, A4 and A5. High-frequency unit attenua- 
tion to 3 db. 



118 



January 1951 Journal of the SMPTE Vol. 56 



quency test film made in accordance 
with American Standard Z22.44. 

2.5: Uniformity of Scanning-Beam 
Illumination: The uniformity of scan- 
ning-beam illumination shall be such 
that the output from the reproducer 
amplifier does not vary more than 
=*=1V2 db when an American Standard 
Uniformity of Scanning-Beam Illum- 
ination Test Film Z22.80 is run through 
the reproducer. 



2.6: Flutter: The flutter introduced 
by the reproducer shall not exceed 
0.25% when using the American Stand- 
ard Flutter Test Film Z22.43. 

Note: This value has been selected as 
the maximum permissible value as 
measured on an RCA Flutter Bridge 
or on an instrument that has been ad- 
justed to give comparable readings. 

2.7: Loudspeaker Attenuation: When 
using the two-way loudspeakers speci- 
fied in the recommendation, it is often 




JOO 400 500 700 OOO 
FREQUENCY IN CYCLES PER SECOND 



2000 JOOO 40005000 7000 10000 



Fig. 3. Recommended electrical characteristic for 16-mm review room re- 
producer employing International Projector Simplex Four- Star Loudspeaker 
Systems Models A and B. High-frequency unit attenuation to 2 db. 




200 JOO 400500 700 OOO 

FREQUENCY IN CYCLES PER SECOND 



2000 JOOO 40002000 7OOO 10000 



Fig. 4. Recommended electrical characteristic for 16-mm review room 
reproducers employing International Projector Simplex Four- Star Loud- 
speaker Systems Model C. High-frequency unit attenuation 2 to 3 db. 



Recommendations for 16-Mm Review Rooms 



119 



advisable to attenuate either the high- 
or low-frequency side of the dividing 
network to obtain equal acoustical re- 
sponse on both sides of the network 



cross-over frequency. Typical values 
of attenuation have been specified with 
each of the recommended response vs. 
frequency curves. 



'). Acoustical Requirements 



3.1: Room Reverberation Character- 
istics: The desirable reverberation time 
of a room is a function of its size. Ex- 
cessive reverberation causes blurring of 



speech and rapidly moving staccato 
music. Where the reverberation time 
in the room is below optimum, an ex- 
cessive amount of sound energy must be 



1 1 
























































-# 




















































40 


Oro-2 


=# 




















































70 
100 


OTO-I 



UK 




































^ 














[= 


300 

500 













m 

















MM 














.^^ 












IS 




\ 




2000 
2300 

J500 


^ 


m 


















































1 


4000 
5000 
6000 


| 

2 i 


EE 






















































-6 


H- 




















































8000 


18 


-M 
























































10 40 50 


71 


D 




00 


^ 


00 


y. 


K> 





X> * 


30 


7( 


X3 







00 


20 


00 


JO 


00 


4C 


O05 


yx> 


TCX 


>o to 


300 





FREQUENCY IN CYCLES PER SECOND 



Fig. 5. Recommended electrical characteristic for 16-mm review room 
reproducers employing RCA Energized Loudspeaker Systems Models PG91, 
PG92, PG117 and PG118. Low-frequency unit attenuation to 2 db. 



m 




















































40 


OB. 
OTo-2 


1 




















































D 100 
] 300 

300 


/'O 




1 


m 


m 





^ 





VM 





. 
















m^rn 


' 


~_ 


. 


s 


^ 


\- 

^-\ 








KXX3 
2000 
?300 

J300 
4000 

3000 





-2 

-4 5 


m 














































jj 


-\ 




8000 
8000 


8 | 
13 
17 



200 3OO 400 500 700 1000 
FREQUENCY N CYCLES PER SECOND 






Fig. 6. Recommended electrical characteristic for 16-mm review room 
reproducers employing RCA Permanent Magnet Loudspeaker Systems Models 
PL240, PL244 and PL246. MI-9458 high-frequency unit attenuation 1 to 2 db. 
VII-9449 low-frequency unit attenuation db. M 1-9-148 high-frequency unit 
attenuation db. MI-9449 low-frequency unit attenuation 1 to 2 db. 



120 



January 1951 Journal of the SMPTE Vol. 56 



radiated and the resultant sound is 
unnatural. 

The optimum reverberation period 
varies with frequency and with the size 
of the room. Figure 8 gives the op- 
timum reverberation time for theaters. 

To summarize, the essential design 
features are : 

1. A minimum volume consistent 
with the required seating capacity and 
proper auditorium proportions. 

2. An auditorium width of from 50 to 
70% of the length and an auditorium 



ceiling height of not more than 40% of 
the length. 

3. The use of nonparallel surfaces; 
in particular, the floor should not be 
parallel to any ceiling section nor op- 
posite side-wall sections parallel, 

4. The use of convex rather than con- 
cave surfaces. In addition, the wall 
and ceiling surface should otherwise be 
broken up so as to thoroughly diffuse 
the sound. 

5. Auditorium absorption character- 
istics to provide the same rate of sound 



M 




















































FREO 
40 


DB. 
Oro-2 


m 




















































100 
300 
500 






1 








5 


























-v^ 




^5 


=-. 


^ 




-\f 


V 
-\ 






1000 
2000 
2500 

J500 
4000 
5000 



-1 

D 

-2 

3 



















































* 




6000 
8000 


7 
-II 

-ia 



2OO JOO 4OO 500 700 1000 

FREQUENCY IN CYCLES PER SECOND 



2000 JOOO 40005000 7000 10000 



Fig. 7. Recommended electrical characteristic for 16-mm review room re- 
producers employing Western Electric Mirrophonic Systems Models Ml, M2 
and M3. High-frequency unit attenuation 2 to 4 db. 




ioo t ooo ipoopoo 

VOLUME IN CUBIC FEET 
Fig. 8. Optimum reverberation time for motion picture theaters. 

Recommendations for 16-Mm Review Rooms 121 



decay in a vertical as in a horizontal 
direction from side to side or from back 
to front walls. 

6. Heavily upholstered seats and 
ozite-lined carpet in the aisles. 

7. Backstage treatment giving a 
negligible amount of reflected or re- 
radiated sound from the backstage into 
the auditorium. 

8. A heavily carpeted stage designed 
for good viewing conditions from the 
front seating section. 

9. Auditorium walls with sufficient 
sound insulation material to prevent 
extraneous noise from entering the 
auditorium. 

10. The projection booth acoustically 
treated with fireproof material and pro- 
jection ports equipped with acoustic 
baffles. 

11. All equipment subject to vibra- 
tion and hum such as arc generators, 
voltage regulators, lighting control 



equipment, etc., acoustically isolated 
from the auditorium. 

12. Air-conditioning equipment of a 
high-volume, low air-velocity type with 
air ducts provided with ascoustic baffles. 

Long, narrow auditoriums, high ceil- 
ings, excessively long and narrow bal- 
cony overhangs, concave focusing sur- 
faces, and large unbroken reflecting 
areas should always be avoided, as 
acoustical faults will always result 
from their use. 

If these recommendations are 
followed, the resulting auditorium will 
give sound (as reproduced on a modern 
two-way equipment) with high intelli- 
gibility, warm, natural screen presence, 
good balance between high and low 
frequencies, uniform loudness level 
throughout the auditorium and the 
proper relative balance between high- 
level music passages and low-level, in- 
timate dialogue scenes. 



4. Test Procedure 



4-1: Inasmuch as the sound-repro- 
ducing equipment in review rooms 
usually is subject to extremely hard 
usage, it is recommended that equip- 
ment purporting to meet these require- 



ments be checked at least once a week. 
It is also recommended that a check 
sheet indicating the results of these 
weekly tests be maintained. 



122 



January 1951 Journal of the SMPTE Vol. 56 



Sound Committee Report 



By Lloyd T. Goldsmith, Committee Chairman 



IT HAS BEEN several years since the 
Sound Committee has reported to 
the Society on its activities and accom- 
plishments. It has been active, how- 
ever, on projects authorized by the 
Engineering Vice-President and the 
following is an account of its work to 
date. 

A subcommittee under the chairman- 
ship of R. T. Van Niman investigated 
the possible advantages of the blue- 
sensitive and lead sulfide types of photo- 
tubes for 35-mm theater and 16-mm 
projector use over the presently used 
red-sensitive phototubes. This is a 
continuing activity being carried on 
with manufacturers of color films; but 
at the present time, only the red-sensi- 
tive phototube is recommended as giv- 
ing the best all-around performance with 
present day black-and-white and color 
sound tracks. Additional data now 
scheduled for collection may provide the 
basis for modifying this statement, how- 
ever. 

Our committee has cooperated with a 
subcommittee of the 16-Mm and 8-Mm 
Motion Picture Committee, which is 
working to establish electrical charac- 
teristics for 16-mm review-room repro- 
ducers. 

We studied and approved proposals 
which lead to the standardization of 
100-mil and 200-mil push-pull sound 
tracks used in recording original sound. 

Considerable correlation has been car- 
ried on to reconcile the Society's pro- 



Presented on October 18, 1950, at the 
Society's Convention at Lake Placid, N.Y. 



posed standards of flutter definition and 
measurement with the proposals of Dr. 
Kellogg's ASA Sound Committee Z-57. 
Agreement has now been reached and 
early ASA standardization will result. 
The original flutter proposals were 
formulated by the Sound Committee 
under the chairmanship of J. G. Frayne, 
who with R. Scoville, has actively fol- 
lowed through with the correlation. 

Our committee aided in the prepara- 
tion and final approval of the Society's 
16-mm Sound Service Test Film Code 
SPSA, which has had wide sale and use 
in testing the performance of 16-mm 
sound projectors. 

The proposed British standards for 
magnetic recording were reviewed and 
comment forwarded to the British Stand- 
ards Institution. 

It was brought to the attention of the 
committee that some recent screen in- 
stallations in theaters resulted in exces- 
sive loss of volume and high-frequency 
response from the screen horns. The 
committee investigated, measured the 
loss of screen samples, and on finding it 
excessive, aided the manufacturer in 
modifying his fabric to reduce the sound 
loss to accepted values. As the War 
Standard Z52.44-1945 "Sound Trans- 
mission of Perforated Screens" had 
never been reviewed and processed as 
an American Standard, the committee 
circulated it to all known screen manu- 
facturers for approval. Their recent 
loss data all met the War Standard, 
and, accordingly, the Sound Committee 
approved the War Standard with minor 
revisions, and the new proposal was pub- 



January 1951 Journal of the SMPTE Vol. 56 



123 



lished in the July, 1950, JOURNAL for a 
90-day trial period leading to its eventual 
adoption as ASA Standard Z22.82. 

The proposed American Standard for 
Acoustical Terminology developed by 
ASA Sectional Committee Z-24 on 
Acoustics was reviewed and suggested 
changes- forwarded to that committee. 

In May, 1948, a Subcommittee on 
Magnetic Recording was set up, with 
G. L. Dimmick as Chairman, and 
charged with the formulation of stand- 
ard sound track dimensions and speeds 
of magnetic recording on 35-, 173^-, 16- 
and 8-mm motion picture film. With 
the help of several task forces assigned 
to specific aspects of the problem, the 
subcommittee prepared for the Sound 
Committee proposed standards which 
are now in the hands of the Standards 
Committee with the recommendation 



that they be published in the JOURNAL 
for six-month's trial and criticism. A 
progress report of the subcommittee was 
given at the 1949 Fall Convention. 

The Magnetic Recording Subcom- 
mittee is about to prepare specifications 
for magnetic test films of the types that 
may be required by industry and sold 
by the Society. At the moment, an 
azimuth film, multifrequency film and 
buzz track appear to be most needed 
and will probably be made available 
first. 

It is anticipated that problems as- 
sociated with magnetic recording and 
reproduction will constitute the major 
part of the committee's work for the 
coming year with particular emphasis 
on standards, test films and television 
sound problems. 



Theater Television Committee Report 



By D. E. Hyndman, Committee Chairman 



DURING 1948 AND 1949, the work of 
the Society's Theater Television 
Committee was devoted to the considera- 
tion of all engineering phases of the use 
of television in motion picture theaters. 
It reviewed the design, construction and 
operation of theater television equip- 
ment, from the standpoint of alterations 
that might be necessary within a thea- 
ter, power supplies, viewing conditions, 
screen brightness, program distribution 
facilities and the like. 



Presented on October 20, 1950, at the 
Society's Convention at Lake Placid, N.Y. 



In June, 1949, the Federal Com- 
munications Commission requested the 
Society along with Paramount Pictures, 
Inc., and Twentieth Century-Fox Film 
Corp. to file a statement concerning the 
allocation of frequency bands for a 
theater television service. This request 
brought to an end the more or less broad 
general consideration that was being 
given to all phases of this work and 
forced the committee to concentrate on 
a statement which would outline in 
specific terms what the industry needed 
in the way of radio frequencies to estab- 
lish a nation-wide theater television serv- 



124 



January 1951 Journal of the SMPTE Vol. 56 



ice. On August 29, 1949, the commit- 
tee filed such a statement with the Com- 
mission. 

It was realized at that time that 
some of the conclusions reached by the 
statement, while based on good engineer- 
ing judgment, could not be backed up by 
actual engineering data. It was also 
realized that such concrete information 
would have to be provided at the time 
of the public hearing if the industry had 
any hopes of having the FCC grant their 
request. 

1950, therefore, has been devoted to 
securing the technical data on distribu- 
tion facilities, which would substantiate 
the 1949 statement. As a means to 
this end, a subgroup was established 
under the chairmanship of George L. 
Beers of RCA. This group is composed 
of theater television equipment manu- 
facturers and representatives of the 
common carriers. They were requested 
to investigate four specific characteris- 
tics of a theater television distribution 
system. The first dealt with the band- 
width required, the second with permis- 
sible signal-to-noise ratio, the third 
with distortion, and the last with the 
compression which could be tolerated on 
such a distribution system. 

RCA agreed to provide the laboratory 
facilities for conducting these tests, 
provided the committee reviewed the 
test methods proposed and gave its 
assistance in interpreting the test results. 
At present, work is in progress on the 



first two of the assigned tasks, namely 
bandwidth and signal-to-noise. The 
subcommittee has approved the test 
methods prepared by Otto Schade and 
is awaiting an opportunity to judge the 
results on a large-screen theater system. 
So far, only limited viewing tests have 
been conducted and these on a small- 
screen direct-view cathode-ray tube. As 
soon as a large-screen laboratory setup 
is made available, it is hoped definite 
conclusions can be reached. 

From the standpoint of practical 
operating problems, a wealth of experi- 
ence will be gained from the actual 
theater installations that have been 
made in recent months. Nine theaters 
in seven cities now have equipment in- 
stalled and are carrying weekly programs 
of various sports events. It is reported 
that before the first of the year, there 
will be 16 theaters so equipped. Since 
both cable and radio facilities are being 
used for program distribution to these 
theaters, much will be learned that 
will assist Mr. Beers' group in reaching 
rapid conclusions. 

The Theater Television Committee 
plans to continue this activity to arrive 
at the answer to the basic engineering 
problem. When this preliminary work 
has been completed, it is anticipated 
that appropriate standards and recom- 
mendations will be set up as the Society 
has done in the past in the field of mo- 
tion pictures. 



D. E. Hyndman : Theater Television Report 



125 



Spring Convention 1951 



APRIL 30 MAY 4 The Society's 69th 
Convention in 

34^ years is also the 69th to be held under 
the able tutelage of Bill Kunzmann, Con- 
vention Vice-President. His staff of 
Chairmen and Vice-Chairmen for the 
Spring Convention have now been ap- 
pointed and by the end of January will 
have completed the general schedule of 
events as well as preliminary preparation 
for details of the program. 

NEW YORK CITY Since the Papers 
Committee Vice- 

Chairman who resides in the city where a 
convention is being held, automatically 
becomes Program Chairman, the choice of 
New York gives the responsibility to Bill 
Rivers. Among other things, he will de- 
velop the details of papers presentation 
along lines suggested by Ed Seeley, Papers 
Chairman, and will prepare manuscript 
copy of the Tentative and Final Programs. 

HOTEL STATLER Technical Sessions 
will be held in the 

Georgian Room on the Ballroom floor of 
the Statler rather than in the Salle 
Moderne as in the past, because increased 
attendance has forced a move to a larger 
meeting room. Headquarters for the 
Ladies' Committee will be in Room 129 
and Conference Rooms 2 or 3 on the 
Mezzanine have been reserved for meetings 
of technical committees throughout the 
week. The Publicity Committee will set 
up shop in Conference Room 8. 

PAPERS Members of Ed Seeley's Pa- 
pers Committee are rounding 
up groups of related papers on subjects 
that are either of special technical interest 
at this time on have been neglected at re- 
cent conventions. As a consequence, the 
program will include several symposium- 
type sessions, each of which will include all 
or nearly all convention papers related to 
a particular topic. Members or guests 
whose interests and whose time are 



limited will be able to derive maximum 
benefit from minimum participation. 

ADVANCE Realistic deadline dates for 
printed material have been 
established as objectives for authors and 
Papers Committee members. The sched- 
ule of sessions, symposium titles, informa- 
tion on tours and major entertainment 
features must be on the editor's desk by 
February 19, so the Advance Notice which 
includes the hotel room reservation card 
can be printed and ready for mailing to all 
members on Monday, March 5. 

AUTHORS By Friday, February 23, 
Bill Rivers must have re- 
ceived from each prospective author, the 
white copy of the 69th Convention Author's 
Form, and two copies of a 50- to 75-word 
abstract for use in preparing original type- 
written copy of the Tentative Program. 
Bill also requires two self-addressed busi- 
ness envelopes (4% X 9^ in. or there- 
abouts), to simplify prompt mailing of 
subsequent Papers Committee corres- 
pondence. 

TENTATIVE Copy for the Tentative 
Program is scheduled to 
be ready for the printer on Monday, 
March 5. Since the convention is to be in 
New York, Vic Allen will arrange for print- 
ing. He expects to have the Tentative ad- 
dressed to all members and in the mail, by 
first class, on Monday, March 26. 

MANUSCRIPTS Each author must 
send the buff copy of 
the Author's Form with his manuscript 
and one full set of illustrations to Vic Allen 
at Society headquarters by March 23. 
The manuscript should be typed double- 
spaced on good bond paper; send the 
original, not carbon copy. Also, send Vic 
the original illustrations, being certain to 
pack them securely. Good photo-engrav- 
ings cannot be made from poor reproduc- 
tions of original art work. 



126 



FINAL The Final Program listing pre- 
sentation times of all papers will 
be ready by Monday, April 23. Each 
author, as well as each technical session 
chairman and vice-chairman, will be noti- 
fied of his schedule in advance so he can 
plan his convention week before leaving 
home. This is an ambitious program that 
calls for active support by all members, so 



give the Papers Committee and the 69th 
Convention Program Chairman a hand. 
If you are preparing a paper, please ob- 
serve these deadline dates. 

If you have any questions, write to Ed 
Seeley or Bill Rivers. Secure Author's 
Forms and Hints to Authors from the 
nearest Vice-Chairman of the 



PAPERS COMMITTEE 

Chairman, Edward S. Seeley, Altec Service, 161 Sixth Ave., New York 13 



Vice-Chairmen 

For New York: W. H. Rivers 

Eastman Kodak Co., 342 Madison Ave., 
New York 17 

For Washington: J. E. Aiken 

116 No. Galveston St., Arlington, Va. 

For Chicago: R. T. Van Niman 
4441 Indianola Ave., Indianapolis, Ind. 

For Los Angeles: F. G. Albin 
American Broadcasting Co., Station 
KECA-TV, 4151 Prospect Ave., Holly- 
wood, Calif. 



For Canada: G. G. Graham 

National Film Board of Canada, John 
St., Ottawa, Canada 

For High-Speed Photography 
J. H. Waddell 

Wollensak Optical Co., 850 Hudson St., 
Rochester, N.Y. 

As soon as the Committee's roster is, 
complete, it will be published with the 
addresses of all members included. 



Atlantic Coast Section Meeting 



TOM H. MILLER of Eastman Kodak Co., 
Rochester, gave an unusually interesting 
talk on photographic color problems before 
the Atlantic Coast Section in New York 
on December 12. A large number of 
colored slides were used to illustrate each 
point of the color problems discussed. 

Mr. Miller first took up the effect of the 
characteristics of the light source on a 
color photograph. Color distribution in 
the source is of secondary importance in 
taking black-and-white pictures, partly 
because the finished picture must neces- 
sarily look different from the original scene. 
The best result is one which is pleasing to 
the viewer. However, a color picture must, 
at least in most cases, reproduce the color 
of the original scene as accurately as 
possible. But here the photographer runs 
into trouble due to variations in illumina- 
tion of the original which may not give 
acceptable pictures even if perfectly re- 
produced by the photographic process. 



The effect of different illumination of the 
subject was illustrated by pictures taken at 
midday and late afternoon of the same 
subject. Using film balanced for daylight 
(midday sunshine) the late afternoon pic- 
tures were quite obviously different and 
in the case of portraits less desirable, al- 
though for special effects the warmer light 
of late afternoon might give just the effect 
the photographer wants. 

The speaker called attention to a variety 
of effects which may occur in outdoor illu- 
mination, so that the color balance may be 
shifted to the yellow, red or blue, depend- 
ing on the subject and atmospheric condi- 
tions. Usually people do not observe these 
changes in illumination as accurately as 
the film does, because of adaptation of the 
eye. This was illustrated by comparison 
of pictures taken under outdoor and indoor 
illumination with the same film, showing 
marked difference in color balance, al- 
though an observer would have said that 



127 



the illumination was white in each case. 
Another characteristic of the illumination 
which is important in color photography 
is specular or diffuse reflection. In general 
saturated colors cannot be obtained by 
diffuse illumination, as for example on a 
cloudy or overcast day. 

Mr. Miller then discussed certain 
characteristics of color photographing ma- 
terials, particularly their inability to repro- 
duce accurately certain colors. Most com- 
mercial materials are balanced to give good 
flesh tones but this does not mean that all 
colors will be perfectly rendered. Due to 
differences in processes the colors not per- 
fectly reproduced will vary from one ma- 
terial to another. Consequently the only 
way to be sure of obtaining desired results 
is to make test exposures on each fabric, 
mat-erial and paint used in a production. 



Even this is not enough since by adapta- 
tion, the eye adjusts itself to the predomi- 
nant illumination and judges adjacent or 
subsequent colors in relation to it. This 
was illustrated by a series of pictures in 
which each varied only slightly in color 
balance from the preceding one. Most of 
them were quite acceptable although the 
range of color balance was very great. 
However, a direct change from one end of 
the series to the other was very noticeable 
and undesirable. This accounts for the 
fact that a color film which is satisfactory 
by itself may not look right when spliced 
between films having considerably different 
balance. The effect of background and 
surrounding illumination on apparent 
color rendition was also shown to be con- 
siderable. C.R.K. 



The 1951 Journal 



As THE SOCIETY GROWS, in size, occasional 
breaks with tradition are necessary to 
accommodate the diverse needs of an ex- 
panding membership. One came a year 
ago when "Television" was added to the 
Society's name, recognizing that its new 
importance had placed television firmly 
alongside motion pictures and synchro- 
nized sound. Another break occurs with 
the change from a single- to a two-column 
format beginning with this issue of the 
JOURNAL. 

Of several reasons for making the 
change at this time, two stand out: First, 
the amount of publishable material ac- 
cepted by the Board of Editors has in- 
creased steadily for four volumes in succes- 
sion, requiring the Editor to exceed his last 
two yearly forecasts of JOURNAL pages to 
be printed. The trend will doubtless con- 
tinue. Second, there has been a steady rise 
in cost of publication resulting from in- 
creased charges for paper, engravings and 
labor. None of these is likely to be reduced. 

Here are two opposing factors one 
highly desirable, the other inevitable 
which have put the squeeze on the Society's 
publications program. 



Under the present circumstances, two 
columns, with reduced margins, held the 
only hope for real savings. Changing the 
trim size by a small amount would have 
helped even more but seemed undesirable 
for the time being. Adopting a different 
printing method could produce no real 
economy because of the small press run. 
Any reduction in quality of the paper 
would have been folly, for the grade used 
in 1950 was about the cheapest available 
and often failed to yield adequate halftone 
illustrations. 

The present format (two 13-pica columns 
retaining the previous typeface, Mono- 
type 8A, set in 9-pt. on 10-pt. body) per- 
mits 37J^% more information to be placed 
on a single 6 X 9 in. page of text. Printing 
and binding economies achieved in this way 
will just about offset certain increased 
charges that became effective in Novem- 
ber, 1950, and others that start with 
January, 1951. 

As a result, each Society dollar spent for 
publications in 1951 will buy as much 
printed information as it would have a 
year ago, even though costs have increased 
substantially during the intervening period. 



128 



Engineering Activities 



Television Film Equipment 

During 1950, the Society joined with the 
Institute of Radio Engineers and the 
Radio-Television Manufacturers Associa- 
tion (RTMA) in a cooperative program of 
standardization and exchange of technical 
data. One result was the combining of two 
Society Committees, the Films From 
Television Committee and the Television 
Film Projectors Committee, with similar 
RTMA projects, under a Television Film 
Equipment Committee. This joint group 
under the enthusiastic Chairmanship of 
Frank N. Gillette, General Precision Lab- 
oratory, met last October during the Con- 
vention at Lake Placid and again on 
January 4 in New York City. 

At the recent meeting, agreement was 
reached on a Proposed Standard for 16- 
Mm Projectors for Television Film Chains 
Operating on Full-Storage Basis. To 
bring the proposal into line with the 
standards policies of RTMA and SMPTE, 
certain proprietary references were deleted 
and it has become largely a detailed 
specification for performance of the equip- 
ment. Tests, methods of measurement, 
references to specific test films and to par- 
ticular test equipments are also included. 

The Committee received favorably sug- 
gestions for area of scan on 16-mm film 
and for picture area in video recording. 
The dimensions considered with the reason 
for being selected will be put into the form 
of a Proposed American Standard. After 
balloting of the entire Committee, the 
proposals will doubtless be published in 
the JOURNAL for a short period of trial and 
criticism. 

Substantial agreement was also reached 
on a proposal to publish recommended 
standard dimensions for slides and 
opaques. The 2 X 2 in. transparent slide 
and the 4 X 5 in. opaque were considered 
most desirable, so the Committee will 
shortly vote upon them. 



16- Mm and 8- Mm 

The 16-Mm and 8-Mm Motion Pictures 
Committee, under the chairmanship of 
Henry Hood, has been exceptionally ac- 
tive. Several of the projects will soon 
appear in the JOURNAL, notably the pro- 
posals on 16-mm and 8-mm film splices 
and on 16-mm projection reels. One 
project, "Recommendations for 16-Mm 
Review Rooms and Reproducing Equip- 
ment," developed by a subcommittee 
under the chairmanship of E. W. D'Arcy, 
appears elsewhere in this issue of the 
JOURNAL. It is being published as an 
interim committee report in the hope that 
sufficient comments will be received during 
the ensuing year to enable the committee 
to formulate proposals for standardization. 

Sound 

At the Lake Placid Convention, the 
Sound Committee, under Acting Chair- 
man, John Frayne, was faced with the 
urgent problem of reaching agreement on 
proposed standards for magnetic sound 
tracks on film. Protracted delay would 
result in incompatibility with the first 
equipments appearing on the market. 
Even more serious is the probability that 
the first manufacturer to produce a mag- 
netic sound projector commercially would 
set the standard. With this understand- 
ing, the Committee hammered out Pro- 
posed American Standards for Magnetic 
Sound Track on 35- and 17^-Mm Film, 
16-Mm Film, and 8-Mm Film and sub- 
mitted them to the Standards Committee 
for its recommendations on publication for 
a 90kiay trial. They will probably be 
published within the next few months. 

The status of the Proposed American 
Standards for Sound Transmission of 
Theater Projection Screens published for 
trial in the July, 1950, JOURNAL, was re- 
viewed. Inasmuch as no adverse criticism 
had been received, it was agreed to submit 
it to the Standards Committee for recom- 
mendations on final approval as an Ameri- 
can Standard. 

In addition, plans were made for in- 
creasing activity on lead sulfide phototubes 
and standards for 16-mm magnetic film, 
coated full width. 



129 



BOOK REVIEWS 



Fundamentals of Acoustics 

By Lawrence E. Kiiisler and Austin R. 
Frey. Published (1950) by John Wiley, 
440 Fourth Ave., New York 16. 499 pp. 
+ 5 pp. appendix + 3 pp. glossary + 6 
pp. index. 163 illus. 5^ X 8^ in. 
Price $6.00. 

This book presents the fundamentals 
underlying the generation, transmission 
and reception of acoustic waves. It was 
prepared as a textbook on the funda- 
mentals of acoustics and is a very usable 
book for this purpose. The illustrations 
are good and each chapter is followed by a 
set of very well chosen problems. 

The first half of the book develops the 
theory of vibration of solid bodies and the 
propagation of sound waves through 
fluids. It starts with simple oscillators 
having a single degree of freedom. In a 
logical manner follow chapters on the 
vibration of strings, bars and stretched 
membranes. The general acoustical wave 
equations for fluids are developed and ap- 
plied particularly to plane and spherical 
waves with various boundary conditions 
including transmission from one medium to 
another. Then follows the fundamental 
theory of the radiation of sound from vi- 
brating bodies of various sorts such as pis- 
tons, vibrating spheres, etc. These prin- 
ciples are applied to Helmholtz resonators 
and acoustic filters. Finally in Chapter 9 
there is a brief but excellent treatise on the 
absorption of sound waves under various 
circumstances. 

The theory developed in the first half of 
the book is applied to direct radiator loud- 
speakers and horn-type loudspeakers. 
Chapter 12 is a discussion of microphones; 
carbon, condenser, crystal, electrodynamic 
moving coil and velocity ribbon. The 
electroacoustical reciprocity theory is very 
clearly presented and applied to the cali- 
bration of these microphones. 

There is a chapter on psychoacoustics 
dealing with the mechanism of hearing, 



loudness, masking, binaural localizations, 
etc., followed by chapters on each of the 
following general fields : architectural acous- 
tics, underwater acoustics and ultrasonics. 

The authors have maintained a very 
good balance between the fundamental as- 
pects of the physics of the problems and 
the engineering applications. Numerous 
references are made to analogous electrical 
problems, but this is not overdone and 
each important equation is derived from 
the fundamental laws of physics. 

It should serve as a very useful text in 
senior college and graduate courses, both 
in physics and engineering classes. Dr. 
Harvey Fletcher, 5 Westminster Rd., 
Summit, N.J. 



Fundamentals of Optics, New 
2dEd. 

By Francis A. Jenkins and Harvey E. 
White. Published (1950) by McGraw- 
Hill, 330 W. 42d St., New York 18. 626 
pp. + 4 pp. Answers to Problems + 17 
pp. index + xi pp. 447 illus. 6 X 9 in. 
Price $7.00. 

This book represents a new edition of the 
authors' well-known Fundamentals of 
Physical Optics, first published in 1937. 
As a physical optics text, it is hard to see 
how this book could have been improved, 
and it is gratifying to find that it has been 
reprinted almost without change in the 
new edition. A few sections have been 
added, covering the quantum nature of 
light and some modern developments such 
as the Twyman-Green interferometer, 
phase-contrast microscopes, interference 
filters, and gratings giving a 'blaze' in one 
order. Each topic has been treated with 
just the necessary degree of detail for 
students' use, and difficult side-issues have 
been carefully avoided. Having read any 
chapter, the reader has the pleasant feel- 
ing that now he knows all about that sub- 



130 



ject. The diagrams are clear, and the 
photographic illustrations excellent. A 
particularly gratifying feature of the treat- 
ment is that mathematics is used only to 
provide a deeper analysis of some physical 
phenomenon which has already been ex- 
plained in a clear qualitative way. Too 
many teachers reverse this process, and 
feel that a mathematical treatment is the 
whole story. The book can be confidently 
recommended as an unusually clear ex- 
position of the nature and properties of 
light. 

The new edition also contains a lengthy 
section (175 pp.) on geometrical optics, 
which justifies the more general title. Un- 
fortunately the method of treatment here 
is not nearly as good as that adopted for 
the physical optics part. Fermat's and 
Malus' theorems, and the dispersion of 
glass, are clearly treated, but they are 
actually physical optics phenomena. No 
less than 52 pp. are devoted to the formu- 
las for conjugate distances and magnifica- 
tion, first for a thin lens, then for a single 
refracting surface, then again for a thick 
single lens, and finally for a spherical 
mirror. Surely it would be simpler, and 
more satisfying to the student, to derive 
the formulas for a general optical system 
defined by its two focal points and two 
principal points, and then to regard thin 
lenses and single surfaces as simple special 
cases. 

It is good to find a brief reference to the 
photometry of optical systems and the 
theory of image brightness. Spectroscopic 
and other prisms are adequately covered. 
The properties of chromatic aberration are 
described clearly, but spherical aberration 
is treated in unnecessary detail. The 
references to coma and the sine condition 
suffer from the usual misunderstandings; 
for example, the term "sine condition" is 
used first to refer to the "sine theorem" 
(Eq. 81, p. 121), but later it is used to 
refer to the difference A/ between the 
focal lengths of a lens for paraxial and mar- 
ginal rays (Fig. 9K). The word "coma" 
is correctly used as a transverse measure 
of an aberration pattern in Fig. 91 (b), but 
in Figs. 9K and 9L, and in Table 9III, 
the same term is used to represent the 
longitudinal difference between the A/ 
curve and the spherical aberration curve. 
Obviously both meanings of the same 



word cannot be correct. The diagrams of 
distortion (Fig. 9T) are misleading, for 
when a lens suffers from barrel distortion, 
all parts of the image are too small, the 
corners being excessively reduced in size; 
likewise in pincushion distortion all parts 
of the image are too large, the corners 
again being excessively so. Figure 9V, (b), 
is incorrect, for a single lens with central 
passage of the light cannot possess any 
lateral chromatic aberration. This is an 
aberration of the chief ray, and will appear 
only where the chief ray has been dispersed 
into a spectrum by eccentric passage 
through a lens. The Huy gens' eyepiece is 
referred to in 9.11, line 1, as an achro- 
matic system; this is, however, contra- 
dicted later in the same paragraph. 
There are two errors in labeling of lens 
cross-sections: in Fig. IOC, the diagram 
shows the Zeiss Topogon, not the Ross 
Wide-angle, and in Fig. 10G, the lens 
shown is the "Varo," not the "Zoomar." 
The Galilean telescope diagram in Fig. 10R 
is incorrect, for the eye is actually the 
exit pupil, and only those rays which enter 
the eye should be considered. The en- 
trance pupil of a Galilean telescope is vir- 
tual and situated at a considerable distance 
behind the eye. 

The book is very well produced, on good 
paper, and beautifully printed. A series of 
useful review problems has been included 
at the end of each chapter. R. KINGSLAKE, 
Eastman Kodak Co., Rochester, N.Y. 



Electrical Engineers 9 Handbook 
-Electric Communication and 
Electronics, Vol. H, 4th Ed. 

Edited by Harold Fender and Knox Mc- 
Ilwain. Published (1950) by John Wiley, 
440 Fourth Ave., New York 16. i-xiii + 
1,564 pp. including approx. 130 tables and 
approx. 1,050 illus. + 54 pp. index. 5^ X 
8Min. Price $8.50. 

This edition has been entirely rewritten 
and enlarged to meet the widening fields 
of communication and electronics. Each 
section is written by an expert in that field 
and is accompanied by a bibliography. 

The twenty-three sections cover a wide 
variety of electronic applications as well as 
fundamental properties of materials and 



131 



circuit elements. Frequency modulation, 
television and radar have been given con- 
siderable space. 

As is the case with any handbook at- 
tempting to cover such a wide field, the 
space devoted to any one subject must be 
small compared to a textbook on that sub- 
ject. In the present volume the editors 
and authors have shown good judgment in 
selecting tables and formulas to which a 
worker familiar with the subject may refer, 
and sufficient description so that one un- 
familiar with the particular subject may 
obtain a good introduction to it. CLYDE 
R. KEITH, 5 N. Terrace, Maplewood, N.J. 



Television, Volume V (1947- 
1948) 

Edited by Alfred N. Goldsmith, Arthur F. 
Van Dyck, Robert S. Burnap, Edward T. 
Dickey and George M. K. Baker. Pub- 
lished (1950) by RCA Review, Radio 
Corporation of America, RCA Labora- 
tories Div., Princeton, N.J. i-x + 458 pp. 
+ 3 pp. summary. 315 illus. 6 X 9 in. 
Price, $2.50, plus $0.20 per copy for post- 
age to countries other than U.S. 

Television, Volume VI (1949- 
1950) 

Same editors and publisher, i-x + 402 
pp. + 20 pp. appendix. 284 illus. 6X9 
in. Price, $2.50, plus $0.20 per copy for 
postage to countries other than U.S. 

Television, Volumes V and VI, are 
respectively the eleventh and twelfth 
volumes in the RCA Technical Book Series 
and the fifth and sixth volumes devoted 
exclusively to television. 

The books are comprised of a compila- 
tion of reprints of articles by RCA authors 
which appeared in RCA Review, RCA 
Licensee Bulletin, Broadcast News, Pro- 
ceedings of the I.R.E., the JOURNAL of this 
Society, Communications, Tektech, Journal 
of the Optical Society of America, Electronics 
and Harvard Business Review. 

In the appendix of Volume VI is given a 
complete television bibliography of tech- 
nical papers by RCA authors for the period 
1929 to 1950. Of the total published 
within the periods covered by Television, 
Volumes V and VI, selected articles are 



reprinted in full, others in summary form 
only, while the remainder are omitted ex- 
cept for their listing in the bibliography. 

The papers are presented in each of 
these volumes in six sections: pickup, 
transmission, reception, color, ultra-high 
frequency and general. Within each of 
these sections, distinct phases of television 
development are covered by three types of 
articles: (1) pure theory and analyses of 
performance factors, (2) new techniques 
and proposed new designs not yet reduced 
to practice and (3) descriptions of new 
equipment, facilities, methods, techniques 
and concrete applications of principles re- 
duced to practice. 

Material of the first type serves as a 
guide for the conception and development 
of advanced television designs of the future. 
An outstanding article by Otto H. Schade 
is entitled "Electro-Optical Characteristics 
of Television Systems." It is so advanced 
and basic as to be of permanent value as a 
text. 

Material of the second type forms a basis 
of designs of tomorrow's improved tele- 
vision. An example is entitled "Standardi- 
zation of Transient Response of Television 
Transmitters" by R. D. Kell and G. L. 
Fredendall. 

Of the third type, descriptions of new 
equipment are published soon after the 
equipment design is completed. Thus, 
such descriptions represent the latest 
equipment as of that date. Examples are : 
"New Television Field Pickup Equipment 
Employing the Image Orthicon," by J. H. 
Roe, and "Development of a Large Metal 
Kinescope for Television" by H. P. Steier, 
et al. Descriptions of most of the major 
equipment and circuit features now con- 
stituting present-day television systems 
may be found in these two, together with 
preceding volumes of Television. 

Television papers by RCA authors are 
highly authentic because the findings are 
the results of intensive and extensive activ- 
ities of the writers in all phases of tele- 
vision, each of whom is a specialist in his 
particular field. 

These volumes contain a wealth of au- 
thentic television information in a concise 
form and they should be included in every 
engineer's television library. FRED G. 
ALBIN, 241 S. Wetherly Dr., Beverly Hills, 
Calif. 



132 



Proceedings of the National 
Electronics Conference, Vol. 5 

Published (1950) by National Electronics 
Conference, Inc., 852 E. 83 St., Chicago 19. 
i-x 4- 581 pp. text + xi-xxi pp. Contents, 
Previous Issues. Approx. 600 illus. + 
numerous tables. 6 X 9 in. Price $4.00. 

This book is intended to serve as a per- 
manent record and handy reference of the 
papers presented at the National Elec- 
tronics Conference in 1949. Since only a 
small group can attend such a conference, 
the publishing of this volume allows all engi- 
neers to receive the benefits of the papers. 
Many papers which are presented at such 
meetings are never published elsewhere, 
consequently this provides the only per- 
manent record of these papers. 

There is not sufficient space here to re- 
view each of the 59 papers included in the 
book. However, it can be said that the 
papers range from basic theory to com- 
ponent design and application. Subjects 
covered include audio-frequency, super- 
sonics, magnetic devices, vacuum tubes, 
circuits, theory of communication, anten- 
nas, television, measurements, computers, 
electronic instrumentation and others. 
One paper, "The Magnetic Cross Value 
and Its Application to Subfrequency Power 
Generation," presented a very interesting 
new magnetic device. The analysis of the 
operation of this device was well presented. 
It is refreshing to find that we are still 
discovering new principles in magnetism, 
one of the older phases of electronics. 

It is unfortunate that no attempt was 
made to group papers on the same general 
subject or to provide an index which would 
facilitate rapid exploration of the volume 
to find everything on a particular subject. 

Since the discussion of papers at such 
meetings frequently adds important in- 
formation it is regrettable that no com- 



ment on the discussions of these papers is 
included. 

In spite of these shortcomings the 
N.E.C. is to be commended for publishing 
its Proceedings, and it is hoped that other 
conferences will soon follow suit. OGDEN 
PRESTHOLDT, Columbia Broadcasting Sys- 
tem, 485 Madison Ave., New York 22. 



Manuel de Sensitometrie (3d ed.) 

By L. Lobel and M. Dubois. Published 
(1950) by Publications Photographiques 
Paul Montel, 189, Rue Saint- Jacques, 
Paris (5 e ). 216pp. 103 illus. 5^ X 7^ 
in. Price 375 francs. 

This elementary book gives the principal 
definitions concerning sensitometric meas- 
urements, some of the properties of the 
characteristic curve, and a review of the 
chief systems used for the definition of the 
negative emulsion speed. It includes a 
chapter on paper sensitometry and the 
choice of printing conditions, and another 
on sensitometry for positive films used by 
projection. 

It also gives information on reversal 
development, including the influence of 
the solvent action and that of the second 
exposure, on intensifying and reducing 
processes, on the control of color photog- 
raphy and the use of the masking method. 

Finally, about forty pages concern the 
elementary principles of sound recording 
and the application of sensitometry to 
sound film. 

As regards the apparatus used hi sensi- 
tometry, the descriptions are very short 
and the authors emphasize the densitom- 
eters designed by Mr. Lobel. 

This booklet, despite a few errors and 
many oversimplifications, should be useful 
to amateurs and beginners in photographic 
science. R. PINOIR, Kodak-Pathe, 30, 
Rue des Vignerons, Vincennes, France. 



Journals Available: The following back numbers of the Journal are available from 
Mr. John R. Bizzelle, 419 West 48 St., New York 19, N.Y.: Oct. 1917, $1.00; 
Apr. 1918, $1.00; Sept. 1931 (2 cys) $.50 each; Nov. 1931, $.50; Jan. 1935, $.50; 
all 12 issues for 1942 at $.25 each; all 12 issues for 1943 at $.25 each; all issues for 
1944 (except Mar., Apr. and May) at $.25 each; all 12 issues for 1945 at $.25 each; 
and Jan., Feb., May, June, July, Aug., Sept. and Oct. 1946 at $.25 each. 



133 



New Members 



The following members have been added to the Society's rolls since those published last month. 
The designations of grades are the the same as those used in the 1950 MEMBERSHIP DIRECTORY. 



Honorary (H) 



Fellow (F) 



Active (M) 



Associate (A) 



Student (S) 



Allen, William H., Commercial Photogra- 
pher. Mail: 721 E. Fayette St., Syracuse 
3, N.Y. (A) 

Barnes, Carl E., Director of Chemical Re- 
search, Arnold, Hoffman & Co., Provi- 
dence, R.I. (A) 

Beibin, Harold, Film Recording Engineer, 
Allen B. Du Mont Laboratories, Inc. 
Mail: c/o Brown, 240 W. 98 St., New 
York, N.Y. (A) 

Besse, Armand, Sales Manager, Perkins 
Electric Co., Ltd. Mail: 9370 St. Hubert 
St., Montreal 12, Quebec, Canada. (A) 

Clark, Walter M., Technical Photographer, 
Northrop Aircraft, Inc. Mail: 2907 Gib- 
son PL, North Redondo, Calif. (A) 

Dare, Doug, Motion Picture Cameraman, 
Sam Hayes Productions. Mail: 600 N. 
Maple St., Bur bank, Calif. (M) 

Darmstaedter, Eric, Vice-President, Reeves 
Equipment Corp. Mail: 10 E. 52 St., 
New York, N.Y. (A) 

Dierken, Kenneth C., American Television 
Inst. Mail: 534 Wellington, Chicago 14, 
111. (S) 

Forbes, Richard B., Hollywood Sound Inst. 
Mail: 1021 Palm Ave., Los Angeles 46, 
Calif. (S) 

Fung, David T., New Institute for Film and 
Television. Mail: 435 W. 123 St., New 
York, N.Y. (S) 

Grossman, Milton B., Electrical Engineer, 
Otto K. Olesen Co. Mail: 10401 Tuxford 
St., Sun Valley, Calif. (A) 

Hughes, Lafayette M., Jr., Producer and 
Director, Hughes Sound Films. Mail: 21 
S. Downing St., Denver, Colo. (M) 

Johnson, Carl M., Head, Technical Informa- 
tion Div., U.S. Navy Electronics Labora- 
tory. Mail: 336 W. Upas St., San Diego 3, 
Calif. (A) 

Katz, Leonhard, Engineer, Raytheon Manu- 
facturing Co. Mail: 19 Ward St., Wo- 
burn, Mass. (A) 

Kenworthy, N. Paul, Jr., 10710 Strathmore 
Dr., Los Angeles 24, Calif. (S) 

Keough, James L., Still Camera, Projector 
and Movie Camera Repairman, Craig 
Movie Supply, and self. Mail: 6548 23d 
Ave., N.E., Seattle 5, Wash. (M) 

Kish, Frank, Photographer, National Ad- 
visory Committee for Aeronautics. Mail: 
4481 W. 51 St., Cleveland 9, Ohio. (A) 

Kusack, William P., Chief Engineer, Bala- 
ban & Katz Television. Mail: Station 



WBKB, 190 N. State St., Chicago 1, 
111. (M) 

Lantz, Donald R., Assistant Director, Dept. 
of Audio- Visual and Radio Education, In- 
ternational Council of Religious Educa- 
tion. Mail: 206 S. Michigan Ave., 
Chicago 4, 111. (A) 

Lewis, Louie L., Chief Engineer, WOI, WOI- 
FM, WOI-TV., Iowa State College, Ames, 
Iowa. (M) 

Matilla, Augusto M., National Supply S.A. 
Mail: P.O. Box 2909, Caracas, Venezuela. 
(A) 

Motts, Gordon H., Supervisor, Still and 
Motion Pictures, Army Field Forces, Bd. 
#4, Ft. Bliss. Mail: 613 Mission Rd., El 
Paso, Texas. (A) 

O'Byrne, Frank E., General Manager, 
Queensway Studios. Mail: 277 Victoria 
St., Toronto, Ontario, Canada. (M) 

Phelan, Charles W., Owner, Films for Tele- 
vision, Inc. Mail: Harbor Ave., Marble- 
head, Mass. (A) 

Reiter, Abraham, Instructor, Audio Engineer- 
ing, University of Hollywood. Mail: 
3808 W. Alameda Ave., Bur bank, Calif. 
(A) 

Robyn, Abe, Sound Technician, Universal 
Recorders. Mail: 850J/2 N. Edinburgh 
Ave., Los Angeles, Calif. (A) 

Shelton, Aaron, Chief Engineer, WSM-TV. 
Mail: 2901 23d Ave., S., Nashville 5, Tenn. 
(M) 

Steadman, Loren L., 2911^ llth Ave., Los 
Angeles 18, Calif. (S) 

Williams, David L., Advisory Engineer, Lamp 
Div., Commercial Engineering Dept., 
Westinghouse Electric Corp., Bloomfield, 
N.J. (M) 

Woodside, Robert L., Sound Technician and 
Mixer, U.S. Air Force, Lookout Mountain 
Laboratory. Mail: 8935 Wonderland 
Ave., Hollywood 46, Calif. (M) 

CHANGE IN GRADE 

Jacobsen, Roger G., Supervisor, Audio- 
Visual Installations, University of Wash- 
ington. Mail: 11350 21st Ave., N.E., 
Seattle, Wash. (S) to (A) 

DECEASED 

Lyons, Thomas T., Projectionist, National 
Theaters Amusement Co., 444 W. 56 St., 
New York 19. (A) 



134 



Obituary 



Albert S. Howell, Chairman of the Board 
of Bell & Howell Co., died at the age of 71 
on January 3 in Chicago. He had retired 
from active service in 1940 from the com- 
pany which he had helped form in 1907. 

He was born in 1879 in West Branch, 
Mich., and became an apprentice at 16 
for the Miehle Printing Press Co. He 
finished high school in night attendance 
and later studied in night classes at the 
Armour Institute of Technology. From 
his experience in his teens repairing 
motion picture cameras, his inventions led 
him to getting patents on 65 photographic 
devices. 

Three of his inventions are credited as 
forming much of the basis for standardiza- 
tion on the 35-mm film width early in the 
1900's and the ensuing rapid progress of 
the industry. His inventions were the 
Bell & Howell film perforator (1908), the 
continuous printer (1911) and a standard 
camera. His first patent was on a 35-mm 
projector and that led to the formation of 
the Bell & Howell Co. 

Mr. Howell was an Honorary Member 
of this Society and he was honored in 1927 
by receipt of the Wetherill Medal from the 
Franklin Institute. He was one of three 




men who have received life membership 
in the American Society of Cinematog- 
raphers, the others having been Thomas 
A. Edison and George Eastman. 



Meetings of Other Societies 



Inter-Society Color Council, Annual Meeting, Feb. 28, Wardman Park Hotel, Washing- 
ton, D.C. The ISCC meeting will consist of three sessions. In the morning a 
Discussion and Business Session will be devoted to presentation and discussion of 
various Color Problem Committee reports, and reports from chairmen of delegates 
from each Member Body. The afternoon session will consist of a symposium of 
reports and demonstrations on Color in Government, a program arranged under 
the chairmanship of Dr. Deane B. Judd. In the evening the Photometry and 
Colorimetry Section of the National Bureau of Standards will hold Open House for 
the group. 

Optical Society of America, Mar. 1-3, Washington, D.C. 

American Physical Society, Mar. 8-10, Pittsburgh, Pa. 

American Physical Society, Apr. 26-28, Washington, D.C. 

Acoustical Society of America, May 10-12, Washington, D.C. 

American Physical Society, June 14-16, Schenectady, N.Y. 

American Physical Society, June 25-28, Vancouver, Canada 

American Institute of Electrical Engineers, June 25-29, Toronto, Canada 

Biological Photographic Association, 21st Annual Meeting, Sept. 12-14, Kenmore Hotel, 

Boston, Mass. 



135 



New Products 



Further information about these items can be obtained directly from the addresses 
given. As in the case of technical papers, the Society is not responsible for manufac- 
turers' statements, and publication of news items does not constitute an endorsement. 




The Hydra Pan Head is a new control de- 
vice available from Hydra Pan Head Co., 
2800 Clearwater St., Los Angeles 39, 
Cain . It is mounted on the tripod or 
tripod head and is hydraulically controlled 
to achieve maximum smoothness and 



horizon parallel. The Hydra Pan Head is 
designed for making murals with still 
cameras and panoramic pictures with 
motion picture cameras using telephoto 
lens. 



Proceedings of the Symposium of 
Improved Electronic Components 

This 236-page, illustrated book contains 
papers by 42 electronic authorities and 
experts in the military, manufacturing and 
engineering fields. The volume is the re- 
sult of a symposium sponsored by the 
American Institute of Electrical Engineers, 
Institute of Radio Engineers and the 
Radio-Television Manufacturers Assn., 



with participation by the U.S. Dept. of 
Defense and the National Bureau of 
Standards. It contains articles on minia- 
turization, printed circuits, glass capaci- 
tors, quality resistors, improved tube de- 
sign and the views of aircraft, naval and 
military experts. The sponsors have ar- 
ranged distribution of the Proceedings at 
$3.50 a copy, postpaid if check accom- 
panies the order sent to: The Tri-Electro 
Co., 1 Thomas Circle, Washington 5, D.C. 



SMPTE Officers and Committees: The Roster of Society Officers was published 
hi the May 1950 JOUBNAL. For Administrative Committees see pp. 515-518 
of the April 1950 JOURNAL. The most recent roster of Engineering Com- 
mittees appeared on pp. 337-340 of September 1950 JOURNAL. 



136 



Image Gradation, Graininess and Sharpness 
in Television and Motion Picture Systems 

Part I: Image Structure and Transfer 
Characteristics 

By Otto H. Schade 



The physical quality of motion picture and television images is determined 
by the transfer characteristic, the standard deviation or signal-to-fluctu- 
ation ratio, and the detail flux-response characteristics of the system. 
The performance of typical systems and system combinations is illustrated 
by examples permitting numerical comparison. The analysis of fluctu- 
ation Ievels("iioise") in photographic processes, based on sampling theory, 
includes an evaluation of the sine- wave frequency spectrum of the devi- 
ation as modified by the "aperture" processes of the system. The sine- 
wave response characteristics of typical apertures are developed as well 
as an accurate method of determining the equivalent "resolving aperture" 
(point image) of practical devices from sine- wave response measurements. 
A new system of rating image-forming devices is thus developed permitting 
precise evaluation and comparison of components as well as of complete 
systems including the eye. Part I discusses the transfer characteristics of 
motion picture and television systems. Parts II and III, to be published 
at a later date, will contain an analysis of signal-to-fluctuation ratios and 
detail contrast. 



INTRODUCTION 

rwiHE QUALITY of images reproduced other, has a longer or different tone scale, 
J_ over a television or motion picture or has greater sharpness. These sub- 
system may be judged by a visual com- jective impressions, however, are often 
parison with respect to three character- difficult to separate. Differences in 
istics: tone scale, graininess and sharp- picture size, brightness, contrast and 
ness. Such a comparison may indicate flicker, to mention only a few variables, 
that one image is more grainy than the may have a considerable effect on a 

visual evaluation of the image charac- 

Presented on April 20, 1950, at the So- * eristics - The eye is not capable of per- 
ciety's Atlantic Coast Section Meeting Arming a quantitative and objective 
at New York, N.Y., by Otto H. Schade, analysis of image properties. It can- 
Tube Dept., Radio Corporation of Amer- not, of course, be used at all to evaluate 
ica, Harrison, N.J. the quality of electrical images or image 

February 1951 Journal of the SMPTE Vol. 55 137 



SYMBOLS 

Note: Peak values are designated by a peak sign over the symbol, B; average values 
by a horizontal bar, B; and rms values by two vertical bars, \B\. 



A Picture area or frame area 

B Luminance (brightness) 

B Luminance of "black" level 

C Contrast range 

D Total photographic density 

D* Density above fog and base density 

AD Density increment or density range 
of photographic image 

d Object distance or viewing distance 

E Exposure (unit: meter candle sec- 
onds) or voltage 

EiE 2 Exposure, index indicating order of 
process 

EI( O ) Exposure on first process determin- 
ing "black" level 

E Bias voltage at "black " level 

F Focal length of a lens 

g, G Small- and large-signal transfer fac- 
tors [Eqs. (8) and (9)] 

/ Current 

/o Current set to black signal level at 
transmitter 

signals in intermediate stages of an 
imaging process, nor can it be used to 
evaluate or predict the effect of changes 
or improvements which are possible and 
are expected to occur upon further 
development of an imaging process. 
Television is a young and complex art 
with a large number of old and new 
variables. It will take some time to 
eliminate temporary defects and attain 
consistently the level of image quality 
of which the system is capable. 

Various evaluations of the quality 
and particularly of the sharpness of 
television images have been reported in 
the literature. A careful subjective 
comparison of the relative sharpness of 
images transmitted over television chan- 
nels with different passbands appears 
to be a most convincing and direct 
method of determining, for example, the 
loss of image sharpness caused by reduc- 
ing a 4.25-mc (megacycles) television 
channel to 2.7 me (present coaxial cables) 
or the gain in sharpness when the chan- 
nel is increased to 6 or 8 me. 

The results of these tests, however, 



K Constant 

n Number of samples, light quanta, 

silver grains, etc., according to 

index 
Operating point, (signals x = 0, 

y.O) 

q e Charge of one electron [Eq. (6) ] 
q Energy of one light quantum (see 

footnote on p. 141) 
S Film speed 
S e Photosensitivity, practical unit: 

microamperes /lumen 
x, y Input and output signal values 

measured from (general) 

7 Gamma at a point of an operating 

characteristic [see Eq. (10a)] 

8 Diameter of a sampling or resolving 

aperture 

d s Lens stop diameter 
e Quantum efficiency (over-all ) 
T Transmittance factor 
t Flux 

depend to a great extent on the subject 
material, the source of image signals and 
the excellence of the system compo- 
nents. It is obvious that the reproduc- 
tion of a subject which does not contain 
fine detail cannot be improved by pro- 
viding a system capable of reproducing 
finer detail. A standard 35-mm motion 
picture film, when compared with the 
original scene, is certainly not a perfect 
source of image signals. 

It is well known that duplication of a 
motion picture by a second motion pic- 
ture process, identical with the original 
process, results in a marked degradation 
of detail signals and sharpness. Sub- 
jective observations in which a 35-mm 
motion picture is used as a source of 
image signals to evaluate a television 
process are, therefore, hardly conclusive. 
The inadequacy of such observations is 
particularly evident when the quality of 
the duplicating process approaches or 
exceeds that of the motion picture 
source, because even an ideal imaging 
process can but reproduce the quality 
of the signal source. Ten years ago 35- 



138 



February 1951 Journal of the SMPTE Vol. 56 



mm motion picture film seemed ade- 
quate for television tests as a source of 
picture signals. Film scanners were then 
ahead of direct-viewing cameras in 
quality but still inferior to the motion 
picture. In subsequent years, however, 
the resolution obtainable with camera 
tubes and kinescopes has increased con- 
siderably. This increase has been ob- 
tained in a large measure by providing 
better operating conditions for the 
tubes. The resolution of the standard 
35-mm motion picture was soon ex- 
ceeded in laboratory tests and larger, 
sharper test patterns and even better 
lenses are required to test the image 
sharpness obtainable with a standard 
television channel of 4.25 me. It is, 
therefore, desirable and necessary to 
employ objective methods and a unified 
approach in an analysis of image quality 
to coordinate and compare the charac- 



teristics of optical, electrical, photo- 
graphic, and visual processes. When 
the photographic process is used as a 
link in a television process it is to be 
expected that the photographic process 
must be adapted to the characteristics 
of the television process and its imaging 
quality may then, by itself, be quite 
unsuitable for direct observation by eye. 
The principles for an objective evalua- 
tion of image quality have been dis- 
cussed in an earlier paper. 1 An evalua- 
tion of practical systems and system 
combinations, such as in television re- 
cording, requires an analysis of existing 
processes in greater detail. The transfer 
characteristics, relative fluctuation levels 
(grain and noise), and the detail signal 
response (resolution, detail contrast) of 
the various system components and 
their combinations will, therefore, be 
treated in that order. 



A. IMAGE STRUCTURE AND THE SAMPLING PRINCIPLE 



A common basis for an analysis of 
image quality is indicated by the fact 
that all images have a structure. When 
an apparently uniform area is examined 
under sufficient magnification, at least 
theoretically, it is found to consist of 
particles or groups of particles arranged 
in a regular or a random fashion. The 
particles or groups are, to use a general 
term, "samples" of energy or matter 
which, according to number, arrange- 
ment and size, determine the image 
quality. The imaging process is funda- 
mentally a sampling process. The light 
flux from a scene, the image flux passing 
through a lens, the electron currents in 
television tubes or amplifiers, are a flow 
of discrete samples. Samples of light 
energy, known as "quanta" of light, are 
emitted from points of an object. A 
small fraction of these samples is col- 
lected and directed by the camera lens 
to form an instantaneous image of the 
light distribution at the object. The 
degree of continuity in the image infor- 
mation is obviously dependent on the 
number of samples; with respect to an 



area, continuity depends on the sample 
density. 

In practical processes the optical 
image is formed on a photosensitive 
material which releases photoelectrons 
when it is bombarded by light quanta. 
This sample-conversion process gener- 
ally reduces the number of samples, but 
it permits their accumulation and stor- 
age. In the television process the elec- 
trical samples can be stored directly as a 
"charge image." In the photographic 
process the photoelectrons combine 
with silver ions in a secondary conver- 
sion process to form submicroscopic 
silver samples, which, in turn, can be 
accumulated and stored as a "latent 
image." Following these processes, 
which take place upon "exposure" of a 
light-sensitive surface, are processes of 
multiplication or "development" in 
which the electron energy or the mass of 
the silver sample is increased by large 
factors to become sufficient for the 
transmission of information and the con- 
trol of light sources for image reproduc- 
tion. 



Otto H. Schade: Transfer Characteristics 



139 



1. Continuity, Sample Number and 
Sample Size 

The energy or mass of the original 
sample is increased by supplying and 
attaching to it a group of "secondary" 
electrons or atoms, thus forming a new 
sample unit. If a sample is visualized 
as a three-dimensional particle, it can 
be seen that the original sample may be 
enlarged in one dimension (height, inten- 
sity) but still retain its original cross 
section; or, it may be enlarged in two or 
three dimensions and thus increase in 
cross section. In a three-dimensional 
development (film) the increase in 
sample size must be limited because it 
introduces an error in sample position.* 
In all cases, however, the development 
must be uniform for all samples. The 
optimum value of the sample size de- 
pends on a number of factors to be 
determined later. The sampling proc- 
ess gives, hi principle, a discontinuous 
picture information. Continuity and 
uniformity of an area are, in a strict 
sense, illusions. These illusions depend 
on restricting variations in sample den- 
sity in an area representing a constant 
level to a threshold value, a certain 
small deviation determined by the 
method of observation. It is' not diffi- 
cult to see that deviations in density 
become larger when the sample number 
per unit area decreases, and that res- 
toration of continuity requires, then, a 
process of filling out the spaces between 
samples and a supply of additional 
samples. It is logical to attach these 
additional samples in equal numbers to 
the original sample units thus expanding 
the size of the unit representing one 
original sample. This process obvi- 
ously does not supply new information, 
but rather causes a fusion of possible 
detail in areas equal to or larger hi 
diameter than the spacing between 
original sample centers. The desired 
uniformity requires in most cases over- 
lapping of the expanded areas. 

* The silver speck forms on the outside of 
a bromide crystal. 



2. Integration of Samples by Low-Pass 
Filters (Apertures) 

A device limiting the observation or 
transmission of detail or fluctuations in 
one dimension (time) by fusing and 
integrating all faster fluctuations is 
known electrically as a "low-pass filter." 
A similar optical device spreading light 
samples in two dimensions (over a small 
area) when point sources of light are 
imaged, is an "aperture." It is well 
known that the shape and area of the 
point image made with a pinhole camera 
are controlled by the size and shape of 
the pinhole aperture. The point image 
itself may be identified as the "sampling 
aperture" of the imaging device. The 
sample aggregate or figure of confusion 
formed by an imaging device and repre- 
senting a mathematical point is thus de- 
fined, in general, as the " sampling aper- 
ture" in the image, and the effect on 
image definition and limiting resolution 
is an "aperture effect" or a two-dimen- 
sional low-pass filter effect. Because of 
their resolution limit, lenses, dot or 
grain structures, mosaics and the eye are 
two-dimensional low-pass niters which 
integrate samples within areas equal to 
their sampling aperture. It is, there- 
fore, unnecessary for an image-reproduc- 
ing process to integrate a dot or line 
structure which cannot be resolved by 
the eye, nor is it necessary to reproduce 
detail which cannot be seen by the eye 
in the final image. The characteristics 
of vision as a sampling process are thus 
needed as a standard of comparison. 

Two-dimensional images are an as- 
sembly of point images produced simul- 
taneously (lenses, printing, etc.), or in 
sequence (facsimile, television) by mov- 
ing one sampling aperture over the image 
area. Uniform coverage is obtained 
with one aperture by "scanning" the 
image area along parallel paths (lines), 
the aperture moving at a constant ve- 
locity in the "horizontal" direction along 
the scanning line and progressing step- 
wise by constant increments in the "ver- 
tical" direction. In this manner two- 



140 



February 1951 Journal of the SMPTE Vol. 56 



dimensional information under the scan- 
ning aperture is translated into one- 
dimensional information: a signal cur- 
rent varying in intensity with time. The 
change in dimensions indicates a quad- 
ratic relation between the diameter 
(and resolution) of the scanning aper- 
ture as a two-dimensional filter and the 
passband of an electrical channel. It 
likewise indicates a change of units and 
quadratic rules for combining optical 
and electrical "aperture" effects. 1 

The foregoing discussion has shown 
the similarity of elements and functions 
which must be performed by an imaging 
system. Before specific characteristics 
are treated, it will be of interest to dis- 
cuss a number of general relationships 
which are readily understood from the 
sampling principle and must be satisfied 
before images of a given quality can be 
produced. 

3. Light Energy, Image Quality and 
Image Size 

The image quality is controlled basi- 
cally by the energy levels obtainable in 
the imaging process. The higher the 
useful level of energy and the larger the 
number of samples, the higher can be the 
image quality obtainable by the proc- 
ess. A given image quality is, there- 
fore, characterized by: (1) the total 
number of sample aggregates in the 
image frame; (2) the relative accuracy 
of sample density and location in the 
frame with respect to the original; and 
(3) the size of the sample aggregate with 
respect to the frame size. Hence, when 
the distribution and the total number of 
samples in the frame area of a television 
image or a photographic image are held 
constant and the sample size or sizes are 
expanded or contracted in proportion to 
the frame size, the quality of the image 
remains constant and is independent of 
the frame size. Not only does the 
quality remain constant but it is also 
obtainable with the same scene illumina- 
tion, depth of field and exposure time by 
maintaining a proper relation among the 



optical parameters. These relations 
may be illustrated by examining the 
photographic process. 

It is known that the photosensitivity 
of the primary process in photographic 
film is, in principle, independent of the 
grain size built into a particular film 
type. To make one grain developable, 
certain numbers of light quanta, photo- 
electrons, and silver atoms are required 
whether the grain is large or small. 
The ratio of the number of grains de- 
veloped in an area to the total number 
of light quanta received by the area is 
the over-all "quantum efficiency" of the 
film process (including development).* 
This quantum efficiency can be deter- 
mined from the normal sensitometric 
curves of density, D, as a function of 
exposures, log E, and a grain count. 

The light flux of one meter candle per 
second, E = 1, of white light represents 
the quantities: 
1 Im/sq m n <?o/sq ni = 

1.3 X 10 16 quanta f/sq m 

The number, n , of light quanta, q , 
incident on 1 sq mm of film surface dur- 
ing exposure time is, therefore : 

no/sq mm = 1.3 X 10 10 E 

(E in meter-candle seconds) (1) 

The number, n s , of silver grains obtained 
with a given value E, depends on the 
spectral response and the degree of 
development, 7, of the film and is given 
by: 

n/sq mm = n D* (2) 

where D* is the film density above the 
densities of "fog" level and film base, 
and n is the number of equivalent grains 
at D* = 1 for the particular film type. 

* It is apparent that the spacing between 
developable centers is an important 
factor. 

| This number is the number of quanta 
in the wavelength range, X = 0.40 to 
0.73 M from a black body at 5400 K, 
which would give one lumen. Radi- 
ation outside this range is excluded be- 
cause it contributes nothing to the 
luminous output. 



Otto H. Schade: Transfer Characteristics 



141 



The effective quantum efficiency, e, is, 
therefore : 

c = n./n = n D*/E 1.3 X 10 10 (3) 

The quantum efficiency of film has its 
highest value in the toe region of the 
transfer characteristic, D = f (log E), 
and decreases for larger values of D or 
E. The total light is given by the 
number of grains in the image frame 
multiplied by the quantum efficiency; 
in conventional units it is the exposure 
multiplied by the film area, A (in square 
meters) : 

Light energy = E A 1m sec (4) 

Fora constant grain number and quality, 
the light energy must remain constant 
when film and grain areas are shrunk or 
expanded in proportion. The expo- 
sure, E, will thus change in inverse pro- 
portion to the frame area, A; and the 
film speed, S = K/E, will change in 
direct proportion to the frame area. 
(K is a constant.) 

An analysis of the optical parameters 
for the exposure of the film furnishes the 
following facts. If a distribution of the 
light flux emitted from object points 
according to Lambert's law is assumed, 
the quantum efficiency of the camera 
lens is expressed by the ratio of image 
flux, if/i, to object flux, ^ , and given by: 



r(*./2d) 



(5) 



for the practical condition, d ^> 6,, 
where r is the transmission factor of the 
lens, 5 8 , the stop diameter, and d, the 
object distance. For a given angle of 
view and object distance the "depth of 
focus" is a geometric factor controlled 
by the lens stop diameter, 6.. The re- 
quirements of a constant quantum input 
[Eq. (4) ] and depth of focus are thus ful- 
filled by a constant lens stop diameter, 
6,, independent of the size of the image 
formed by the lens. The focal length, 
F, of the lens must be changed in propor- 
tion to the image diagonal or the square 
root of the area to maintain the viewing 
angle, F oc A 1 / 2 ; and the //number of 
the lens, therefore, also changes as the 
square root of the image frame area, A. 
Finally, the lens resolution in lines per 
millimeter must change in proportion to 
I/A 1 / 2 , i.e., it must be inversely propor- 
tional to the //number, which is theo- 
retically true. The relations of the vari- 
ous parameters for constant image qual- 
ity are summarized in Table I. 

A given image quality (including 
depth of focus) requires a certain lens- 
stop diameter and a scene illumination 
which is determined by the quantum 
efficiency of the film process. The 
image quality is theoretically independ- 
ent of the size of the image as long as 
the relations in Table I can be fulfilled. 
The diffraction of light sets a lower limit 
to the //number at //0.5 for refractive 



Table I. Requirements for Constant Image Quality in a Frame Size of Area A. 



Image Properties 



Lens Properties 



Photosensitive Surface 
and Signal Development 



Light flux 
Graininess 


= constant 
= constant 


Focal length 
Lens 


oc AH 

= constant 


Quantum efficiency 
Conversion efficiency 


= constant 






diameter 




(Signal develop- 


= constant 










ment) 




Tone range 


= constant 


//number 


oc AY* 


Sample number and 


= constant 










distribution 












(grain) 




Sharpness 


= constant 


Resolution 


ocl/AH 


Sample diameter 


oc AH 


Viewing angle 


= constant 






Resolution 


oc I/AH 


Depth of field 


= constant 






Gamma 


= constant 










Speed rating 


oc I/A 



142 



February 1951 Journal of the SMPTE Vol. 56 



lenses. The smallest practical frame 
dimension, however, is limited to larger 
values by mechanical tolerances, difficul- 
ties in design and correction of lenses, and 
difficulties in the manufacture of films 
having adequate grain sizes, distribu- 
tion and uniformity. 

The question whether a 16-mm mo- 
tion picture film process can be equal in 
quality to a 35-mm process for identical 
lighting conditions has in principle, 
been answered in the affirmative. It 
remains to be shown by analysis whether 
lenses and film of adequate characteris- 
tics are available. 

The relations given in Table I are 
valid also for the television process. 

The fundamental independence of pic- 
ture quality and image size is demon- 
strated by a variety of kinescope and 
camera-tube sizes. Mechanical toler- 
ances, insulation problems, heat dissipa- 
tion, grain sizes, current densities and 
other technological difficulties impose 
limits on the size reduction of practical 
tubes. The fact that television images 
utilize a single image surface to generate 
or reproduce "live" images introduces 
a number of difficulties which are not 
found in motion picture systems. The 
screen of a small kinescope for theater 
projection, for example, must be capable 
of dissipating continuously the total in- 
put power. A motion picture frame, on 
the other hand, is exposed to the projec- 
tor light flux and heat for only ^ 4 sec. 
Small defects or dust particles on a 
camera-tube surface are permanently 
visible and present a serious problem; 
similar defects in each frame of a pro- 
jected motion picture can hardly be 
noticed for statistical reasons. 

A simple comparison of the film proc- 
ess with the electrical process of tele- 
vision can be misleading in various re- 
spects because of differences in the 
low-pass filter response or "aperture 
response" of an electrical channel which 
must be considered when the require- 
ments for equal performance are evalu- 
ated. 



To equal the quality of a 35-mm 
motion picture negative a television 
camera tube such as the image orthicon 
must be capable of converting light 
quanta into useful electrons with an 
equal over-all quantum efficiency. 
Operation with a light range in the order 
of 30 to 1 reduces the electron storage in 
the tube from its short-range value of 
unity by a factor of approximately 2 in 
the low-light range. Absorption of 
photoelectrons by a mesh electrode and 
the addition of a fluctuation level from 
an electron-discharge beam, which may 
be likened to a high "fog" level, require 
a further increase of 3 to 1 in exposure. 
To compensate for these inefficiencies, 
the primary quantum efficiency of the 
photocathode of the tube must be in the 
order of 6 times the over-all efficiency 
of the film process. A primary quan- 
tum efficiency of 100% means that one 
electron charge, q e , is emitted by one 
light quantum.* With the electron 
charge 

q e = 1.6 X 10 ~ 19 coulombs (6) 

a quantum efficiency of 100% results 
in a photocurrent : 

/ = 1.3 X 10 16 q e = 2080 /*a/lm (7) 

The quantum efficiency of 1% is, hence, 
obtained with a photocathode sensi- 
tivity, S. ^ 20 Ma/lm. 

When this quantum-efficiency value is 
divided by six to obtain the equivalent 
over-all quantum efficiency of an image 
orthicon with S e = 20 Aia/lm, the result 
is 0.16%. This value is in the same 
order as that of the fastest film types 
with normal development. Photosensi- 
tivities equal to and higher than the 
above value are obtained consistently in 
commercial tubes and there is evidence 
that much higher values will be obtained 
in the future. 2 



* The spacing factor does not appear in 
this conversion because the photo- 
cathode of the image orthicon is a con- 
tinuous photosensitive surface. 



Otto H. Schade: Transfer Characteristics 



143 



B. TRANSFER CHARACTERISTICS 



1 . Tra nsfer Fac tor sand Ga m ma 

The relation of "sample" numbers or 
sample densities of the output and input 
flux of an imaging device or system is 
described by transfer characteristics. 
A truthful and undistorted reproduction 
of light values by an imaging process re- 
quires that the intensities and ratios be- 
tween light values in the object be 
duplicated in the image. The corre- 
sponding over-all transfer characteristic 
in linear coordinates is a straight line; 
the system response is linear. This ideal 
performance can be obtained by a com- 
bination of components having linear or 
nonlinear transfer characteristics. In 
practical processes the transfer of large 
light ranges is limited at the low-light 
end of the range by fluctuations due to 
lack of samples or by a light "bias." 
At the high-light end of the range it is 
limited by saturation effects because of 
limitations in the sample supply or 
storage capacity of system components. 

The graphic representation of transfer 
characteristics in linear or logarithmic 
coordinates is a useful step in evaluating 
and combining their properties. The 
transfer characteristics of electron tubes 
are usually plotted in a linear-coordinate 
system which is convenient for evaluat- 



ing the effects of constant additive or 
subtractive levels, "biasing" potentials 
or currents, rectification effects, distor- 
tion, and the transfer factors (signal 
ratios) for large and small signals. 

The nonlinearity of the characteristics 
encountered in image "transducers" 
such as photographic film, television 
camera tubes and kinescopes are par- 
ticularly evident when the characteris- 
tics are plotted in linear coordinates, as 
shown in Fig. 1. In many cases a 
signal-conversion process has a transfer 
characteristic following a law of dimin- 
ishing returns such as an exponential 
characteristic or a power law, y = x"Y, 
where the exponent, y, is smaller than 
unity. Either of these characteristics 
may be modified by secondary effects to 
produce characteristics of the type 
shown in Fig. 2, and exemplified by the 
eye, photographic film (density, D, 
versus exposure, E, in Fig. 1), the 
iconoscope, and image orthicon. It is 
evident that characteristics of this type 
can cover a larger range of input-signal 
energy with a given sample number and 
storage capacitance than a linear charac- 
teristic. The "compression" and the 
transfer of incremental signals by fewer 
samples requires, however, a subsequent 




Fig. 1. Photographic 
transfer characteris- 
tics in linear co- 
ordinates. 



144 



RELATIVE EXPOSURE (E,) 

February 1951 Journal of the SMPTE Vol. 56 



re-expansion of signal or light values 
by a transfer characteristic with a higher 
exponent (y > 1, see Fig. 3) in order to 
restore over-all linearity. (Compare 
curve of M.P. process in Fig. 1.) A 
good balance of gradation values de- 
pends on maintaining the ratio, g/G, of 
the incremental signal transfer factor 



g = dy/dx 



to the large signal transfer factor 

G = y/x (9) 

substantially constant over the trans- 
mitted light range. The signal values, 
x and y, are measured from the operat- 
ing point 0.* 

A plot of an operating characteristic for 
unidirectional signals in logarithmic co- 
ordinates will, of course, never show the 
operating point, 0, which is located at 
the origin of the coordinates. 

It is readily shown that the transfer 
ratio, g/G, is equal to the exponent, y, 
when the operating characteristics are 
expressed as a power law. When 



* It is well known in electron-tube engi- 
neering that the operating point can be 
located by an electrical bias anywhere 
on the transfer characteristic of an am- 
plifier stage without transmitting the 
d-c component at the operating point. 
For unidirectional signals the operating 
point of an electronic amplifier normally 
does not coincide with minimum signal 
values but with the average values of the 
signal unless a d-c restorer or black-level 
setting circuit is used. The operating 
point of photoelectric or electrooptical 
transducers such as camera tubes or 
kinescopes cannot be moved along their 
static transfer characteristic because it 
is impossible to eliminate a steady light 
bias, X( ) or y( ), which remains as a 
minimum exposure, Ei( o) , or as a lumi- 
nance, B , in the optical input or output 
signals. In these cases, the operating 
point, 0, can be moved only along the 
electrical coordinate (?/-axis (/) for camera 
tubes, x-axis (E) for kinescopes), the 
optical coordinate of the operating point 
remaining at zero value. 



plotted in logarithmic coordinates, the 
slope, rf(log ?/)/rf(log x), of the charac- 
teristic is equal to the transfer ratio, 
g/G, because 






d (log x) = - logio d x 



(8) and 



d (log y) = dj 
d (log x} d s 



g/G (10) 



for the condition that the operating 
point is the origin of the coordinate sys- 
tem. 

In photographic terminology the 
maximum slope of the film transfer 
characteristic, D = f (log E), has been 
termed the "gamma" of the film. The 
slope or gradient at any point may be 
defined as the point gamma, y. Be- 
cause of the identity, D = - log r 
(r = transmittance), the point gamma 
equals the log-log slope : 

Y = d(-logr)/d(logtf) (11) 

It is, therefore, suggested that Eq. (10) 
be adopted as a general definition of the 
point gamma, i.e., for the condition 
x = 0, y = at the operating point 0: 



= y (lOa) 



d(\og x) 



This definition agrees in every respect 
with Equation (11) and requires that 
the operating point 0, be placed always 
at the origin of the coordinate system. 
The point gamma, y, of a sensitometric- 
film curve can thus be obtained from a 
linear plot of the transmittance charac- 
teristic, r = f (E), shown in Fig. 1, by 
determining the transfer ratio, g/G = 






, or from a logarithmic plot as 



the slope d(log - r)/d(log E)* These 



* It is noted that the gamma of a negative 
or positive film is a negative quantity, 
while a reversal film has a positive 
gamma. In use the sign is usually 
neglected. 



Otto H. Schade: Transfer Characteristics 



145 



RBITRARY UNITS 




Fig. 2. Signal compression due to Fig. 3. Signal expansion due to 

transfer characteristic following transfer characteristic following 

power law with exponent less than power law with exponent greater than 

unity. unity. 




0.001 



-2. 



-1.5 -I ^- 

LOG X = LOG (E/ E ) 



146 



0.01 2 4 6 8 O.I 2 4 6 8 1.0 

RELATIVE GRID-SIGNAL VOLTS (E/E) 
February 1951 Journal of the SMPTE Vol. 56 



Fig. 4. Effect of 
additive or sub- 
tractive con- 
stants on point 
gamma of kine- 
scope transfer 
characteristics. 



i 1 r~n 1 1 rn 

EXPERIMENTAL IMAGE ORTHICON 

FACE PL ATE =4 1/2 INCHES 

MESH = 500PER INCH 

SPACING=0.00035 INCHES . 

TARGET CAPACITANCE= 1300 jJWt 

FIELD = 70GAUSS 

TARGET POTENTIAL=2VOLTS 



PHOTOGRAPHIC 

STEP TABLET 

IN WHITE FIELD 



PHOTOGRAPHIC 

STEP TABLET 
IN DARK FIELD 



RANGE OF E | CAUSING 
ELECTRON REDISTRIBUTION 

(ELECTRON FLARE) 
"KNEE" POINT 
(APPROXQ 




0.001 



IMAGE ILLUMINATION CEO -FOOT-CANDLES 



Fig. 5. Dynamic transfer characteristics of an image orthicon 
containing d-c signal components. 



methods of determining the point 
gamma apply to all types of transfer 
characteristics. Additive signal con- 
stants, such as a superimposed light 
bias, B 01 on a film or a kinescope screen 
alter the value, G, the shape of the 
characteristic curve in log-log coordi- 
nates and, hence, 7. 

The effect of additive or subtractive 
signal constants on the point gamma 
may be demonstrated on the kinescope 
transfer characteristic shown in Fig. 4. 
The theoretical electrical characteristic, 
0, of the kinescope has a gamma of 
three. Due to phosphor saturation or 
loss of current in electron guns, the 
gamma, 7, of the high-light curve sec- 
tion is reduced. Scattered light or 
ambient illumination represents an 
additive signal constant, B , in the 
order of 1 to 2% which, when added to 
all jB-values, produces the dynamic 
characteristics (curves 1 and 2) having a 
toe. 

In certain cases (television recording) 
it is desirable to increase the gamma in 
the low-light range. The grid-bias 
voltage for zero signal is then moved 



above the cutoff value by a voltage 
bias, E . This displacement of the 
operating point along the ic-signal co- 
ordinate requires that the characteristic 
be redrawn by subtracting E from all 
signal values to furnish the operating 
characteristic curve 3. The light bias, 
B , however, which is caused by E , 
cannot be subtracted and remains in the 
optical output signal. The increase of 
7 due to E in the low-light range is 
obvious from the drawing. 

2. Transfer Characteristics of 
Television Camera Tubes 

A family of image-orthicon transfer 
characteristics 1 plotted in semilogarith- 
mic coordinates and containing d-c 
components caused by optical and elec- 
tron "flare" is shown in Fig. 5. Meas- 
urements with various image-orthicon 
types have shown that the shape of the 
transfer characteristics is determined by 
the operating mechanism of this type. 
The characteristics of different tubes 
differ, therefore, mainly in the numerical 
values of the scales. These differences 
are determined by the target capacitance 



Otto H. Schade: Transfer Characteristics 



147 



10 

06 
*\ 0.4 

Z 

o 
z OJ 
g 0.06 

> 0.04 
> 

u 0-02 

flC 

0.01 



1 1 II 
IMAGE ORTHICOf 
TARGET BIAS + 
" CUTOFF . 


M(C 
2V 


LOSE 
ABO 

1. 


SP 

^E 

X 


*,c 


:D l 


^- 


"^ . 


^ 


^ 


^ 


?*' 


^ 


x 1 " 


X 


^ 


^ 














'/ 


/En. 


t ~"72 


1 


'^ 


/ 


^ 


^ 


/ 
















/ 


//' 


/ 


2.> 
/ 


V 


X! 

/j 


> >( 

^O.( 


1 


















/ y 


/ 


/; 


/ 


/ 


X 






















/ 


'/ x 


X 


/ 


/ 
























/ 


'/ 


/y/ 


/ 


7 


























7 


/ 


/ 






























006 001 0.02 0.04 O.I 0.2 0.4 1 



RELATIVE EXPOSURE (Ej/1,) 
Fig. 6. Dynamic transfer characteristics of close-spaced image orthicons. 



secondary-emission ratios, and photo- 
sensitivity of the particular type. In 
normal operation the "flare "-current or 
d-c black-level current, /, is subtracted 
by setting the video black-level signals 
to zero. The resultant transfer charac- 
teristics, letting 1 = 0, redrawn in 
logarithmic coordinates (Fig. 6), are 
typical dynamic operating characteristics 
for the image orthicon.* The dashed 
curves represent the condition of over- 
exposure which produces undesirable 
edge effects. 

The change of gamma in the transfer 
characteristics of television camera tubes 
with increased exposure comes from 
two causes: (1) diminishing collection 
of secondary electrons (photoelectrons 
in the iconoscope) from high-potential 
target areas; and (2) scattering of the 
uncollected electrons or of flare light 
over a portion of or the entire target 
area. 



* The relative exposure is specified by the 
ratio of the high-light exposure, &, 
to the exposure, E knw , where the high- 
light values are located at the shoulder 
or "knee" of the transfer characteristic. 



The first effect is desirable and similar 
to the expedients used for obtaining 
low-gamma film characteristics, i.e., 
incomplete development. The second 
effect, light- and electron-flare, is unde- 
sirable as it may introduce edge effects, 
level variations and a threshold for low- 
light values. Electron "flare" in image 
orthicons can be particularly undesir- 
able because of its nonuniform distribu- 
tion. 

The d-c black-level current, 7 , in- 
creases with exposure because of optical 
and electron "flare." The "black" 
signal level is, therefore, actually a gray 
signal value (see Fig. 5) but may be re- 
set at the transmitter to a perfect black 
signal value by subtracting the d-c sig- 
nal component. The signal range which 
can be seen on the kinescope screen de- 
pends, of course, on the kinescope 
brightness range and the over-all gamma 
of the system. 

The changes in the gamma of the 
camera-tube characteristics obtained by 
varying the exposure are shown in Fig. 
7. A high exposure (broken-line curves 
in Fig. 6) reduces the gamma in the high- 



148 



February 1951 Journal of the SMPTE Vol. 56 



3.0 
2.0 

1.0 










~ 


IMAGE ORTH 
JET BIAS: 2 VOLTS 


r ON 










TARC 


ABOVE CUTOFF 














^ 






_E,/E KB ..= 


0.6 
0.4 

0.2 
0.1 


















"^^ 

N 


^^^ I.J 

^ ^ ^-^ 




















^^ 23 ^s^ 




















31 ^ 




















47 "^x 










































6 8 


2 468 





1.0 



RELATIVE EXPOSURE (E,/E,) 
Fig. 7. Point gamma of transfer characteristics given in Fig. 6. 



Figs. 8 a,b and 9 a,b are on plate pages 170 and 171. 



lights and increases electron redistribu- 
tion which causes strong local flare and 
distortion of black levels, overempha- 
sizes edges and defects, and coarsens fine 
gradations. (See Figs. 8a and 8b.) A 
lower normal exposure (curve 1.2 or 
2.3, Fig. 6) has a finer natural tone scale 
and if desired can be corrected elec- 
trically to a lower gamma. (See Figs. 
9a and 9b.) These characteristics are 
duplicated by the commercial image- 
orthicon types 3 with 3-in. face plates. 

Rendition of gradation in fine detail 
(texture) requires a low level of random 
fluctuations and spurious signals, and 
good but not overemphasized resolution. 
It is well known that spurious detail 
signals such as those caused by dust par- 
ticles, small scratches, or a collector 
mesh structure go unnoticed in larger 
storage surfaces (films, targets) but can 
cause considerable difficulty in small 
image surfaces because of the high mag- 
nification on the final viewing screen. 
A larger target surface results in higher 
resolution and better texture because of 
the reduction of defects and mesh struc- 
ture in proportion to image size. It 



also results in a lower fluctuation level 
because of increased storage capacitance. 
A larger target surface can be combined 
with a small optical image on the photo- 
cathode of the image orthicon by elec- 
trical image magnification. (The effect 
of a mesh structure will be discussed 
further under resolution.) 

The dynamic transfer characteristics 
of the iconoscope are shown in Fig. 10 
and are replotted in log-log coordinates 
in Fig. 11. These characteristics may 
again be regarded as typical when scalar 
values are considered as relative values 
which may vary for different tubes and 
tube sizes (1848 or 1850). As in the 
image orthicon, the combined effect of 
optical and electron flare (redistribution 
of electrons) causes the fundamental 
effect of raising the black-level signal, 
although the d-c level signal is normally 
not transmitted. Because of the fairly 
uniform distribution of the flare light 
and "flare-electrons" over the target, 
the black-level current depends more 
accurately on the average illumination, 
EI of the image surface. Subtraction of 
the d-c level signals, 7 , furnishes the 



Otto H. Schade: Transfer Characteristics 



149 



in 

Ul 

|0.3 
1 

y 

2 


ICC 


NOSCOPE-TYPEI848 
PH OPTIMUM BIAS LIGHT 
CONTINUOUS EXPOSURE 
RGET CAPACITANCE = 1000 


















/ 


AN 
TA 


>> 


w 














^v 


jf 








































E 


EAM 

o 


CUF 
2t 


\R 

a 


E!> 


JT = 


^ 






























,f 






































f 








AL CURRENT ( 

> < 

r 


























/j 


/ 




^ 


x^ 




























/ 








X 














E,^5 








^ 


^v 






^ 

























1 ' 




Jf 


X 


BE 


A 


M CU 
0.11 


RREN 


T = 




z "' 

O 

8 
o. 






















/$ 


? X 
X 






























s 


^ 


/v x 




























^ 


- 


^ 


'6 


V 1 


x 


















5. * 




- 
< 


J 

( 


= 

'o 


10 ^ 

I 2 


^ 

. 


\ f. 


~ 


r 


o : 


' 




$ ( 


V 


o s 


i 


t ( 


e 



IMAGE ILLUMINATION {E | ) -FOOT-CANDLES 



Fig. 10. Dynamic 
transfer characteris- 
tics of iconoscope 
containing d-c sig- 
nal component. 



operating characteristics such as E\ = 
5; I = 0.01 in Fig. 11. 

Although desirable in its effect of 
increasing the exposure latitude of the 
camera tube, the mechanism of incom- 
plete electron collection and subsequent 
redistribution can cause excessive flare 
and nonuniform levels (shading, dark 
spot) . 

The orthicon is a camera tube having a 
linear transfer characteristic (7 = 1) 
and is free of redistribution effects. 
The linear relation of signal current and 
exposure, however, requires operation 
with high charges and large currents to 
accommodate the highlights in normal 
scenes. In practical designs difficulties 
in maintaining adequate resolution with 
high-beam currents limit the maximum 
useful storage capacitance and seriously 
impair the ' 'signal-to-noise" ratio in the 
medium- and low-light region of the 
kinescope image when the system is 
corrected to approach a linear over-all 
transfer characteristic (discussed in 
Part II). 

Storage camera tubes with substan- 
tially linear response such as the orthi- 
con, the British C.P.S. Emitron, and cer- 
tain types of Vidicons 2 (an orthicon 
type with photoconductive target) have, 
therefore, a short exposure latitude re- 
quiring low-contrast scene lighting, sub- 



dued highlights and critical control of 
camera-tube exposure. 

An analysis of fluctuation levels in 
television images (Part II) points out 
that a natural and constant gamma in 
the charge storage mechanism of the 
order of j = 0.5 (not by redistribution) 
overcomes the above limitations; the 
tube operation remains within the bound- 
aries of practical signal development 
by electron beams. The characteris- 
tics of such a camera tube are, therefore, 
of interest for comparison with com- 
mercial tube types. Its transfer charac- 
teristic in log-log coordinates is a 
straight line with the constant slope, 
7 = O.5.* 

The operating characteristics of camera 
tubes are sections of the dynamic trans- 
fer characteristic extending upward 
from a minimum exposure value, Ei (0) 
and corresponding video signal current 
7 . The zero point of the operating 
characteristic is, therefore, a function of 
the scene contrast range. The value, 
Ei w , depends further on the interpreta- 
tion by the camera operator. His 

* The author was informed some time 
ago by Dr. A. Rose of the RCA Lab- 
oratories at Princeton, that one-half 
power-law transfer characteristics could 
be obtained with photoconductive tar- 
gets. 



150 



February 1951 Journal of the SMPTE Vol. 56 




0.001 




ICONOSCOPE 



0.2 



0.4 0.6 1.0 2 4 6 8 10 20 

RELATIVE TARGET ILLUMINATION (E,) 



40 60 100 



Fig. 11. Iconoscope transfer characteristics of Fig. 10 replotted 
in log-log coordinates. 



Fig. 12. Operating 
characteristics of 
camera tube having > 
a transfer charac- <* 
teristic following a 
one-half power law. 




00! 



0.01 1 A 

RELATIVE EXPOSURE (E|/E,) 



Otto H. Schade: Transfer Characteristics 



151 



TRANSFER CHARACTERISTICS 
POINT GAMMA 




0.001 



2 468 2 468 

0.01 0.1 1.0 

RELATIVE VIDEO SIGNAL VOLTS (E/E) 



0.01 



Fig. 13. Operating 
characteristics and 
point gamma of 
kinescope. 



NEG. FILM. 5231 PLUS X =0.8) 
NORMAL DEVELOPMENT 



= 0.8 / 



LOG EXPOSURE (LOG E,") 



Fig. 14. Transfer characteristic of Plus X 
negative film. 



; 
| 2 

I 

-a 

1 ' 
g 






-ILM 
NOR 


53 

vlAL 


)2 R 
DEV 


ELEASE 
ELOPMEI 


POSITIVE 
4T 


-f 


















/ 


















t 


f 


















/ 
















/ 


f 


'\ 


^ 












/ 




1 




^ 


^ 








^ 


/ 


i 
















/ 




A 












/ 




/ 
















/ 




/ 














/ 




/ 
















/ 


/ 
















/ 


x 


















- *""^ 





















LOG EXPOSURE (LOG E,) 

Fig. 15. Transfer charac- 
teristic and point gamma 
of fine-grain release posi- 
tive film. 



152 



February 1951 Journal of the SMPTE Vol. 56 



decision of letting a certain exposure,, 
E\ w represent a "black level," and 
setting the corresponding video cur- 
rent, 7 , to zero level, corresponds to the 
subtraction I' = I I and the expan- 
sion of this value by gain adjustment to 
a normal signal amplitude by l/(7 7 ). 
Camera-tube operating characteristics 
are, therefore, derived from the primary 
dynamic characteristic by changing the 
video current, 7, at any given value, E\ 
to the operating current : 

7' = (7 - /.)/(/ - 7 ) (12a) 
A reduction of the exposure range and 
setting, I = 0, causes, hence, an in- 
crease in gamma, 71, of the camera-tube 
transfer characteristic as illustrated in 
Fig. 12 by the operating characteristics 
of a camera tube with a normal gamma 
of 0.5. 

3. Kinescope Operating Character- 
istics 

Kinescope dynamic transfer charac- 
teristics obtained with picture modula- 
tion are constructed from the static 
transfer curve S (taken from published 
technical data for kinescope types in the 
RCA Tube Handbook) shown in Fig. 13, 
by adding the flare and ambient light 
bias, B , which is determined by optical 
conditions in tube and viewing room. 
A maximum measured screen contrast 
range, C = 100, for example, in a nor- 
mal image furnishes B = 0.01.B and the 
dynamic characteristic curve in Fig. 
13. The operating characteristic of the 
kinescope is a section of the dynamic 
characteristic and can be adjusted to a 
variety of values by the black signal 
level setting, E , at the receiver. The 
conditions, EJE = 0.05 and 0.09, 
shown in Fig. 13, represent zero-signal 
settings close to subjective black. The 
corresponding operating characteristics 
are constructed by expanding the signal 
range, E E , of curve to unity. 
The signal voltage E' for these corre- 
sponding operating characteristics are 
determined from: 

E' - (E - E.)/(& - E.) (12b) 



For the range, EJE = 0.09, and the 
signal voltage, E = 0.14 at point P, for 
example, Eq. (12b) furnishes the ex- 
panded value, E' = 0.055 for point P'. 
Figure 13 shows that kinescope operat- 
ing characteristics have a lower maxi- 
mum gamma, y = 2.2 to 2.3, than the 
original static characteristic, 7=3. 

4. Motion Picture Film Character- 

istics 

The transfer characteristics of Plus X 
negative film (5231) and type 5302 fine- 
grain release positive film are shown in 
Figs. 14 and 15.* The characteristics 
were measured with substantially paral- 
lel light on II B sensitometer step 
exposures. The developed films ob- 
tained from the Motion Picture Film 
Dept. of the Eastman Kodak Co., New 
York, N.Y., received standard motion 
picture processing (spray process on 
negative, deep tank on positive by 
DeLuxe Film Laboratories, New York, 
N.Y.). 

5. Over-All Transfer Characteristics 

The combination of several transfer 
characteristics in an imaging system re- 
sults in a curved or /S-shaped over-all 
characteristic. The characteristics of 
an image orthicon (Fig. 6) (I = 0) in 
combination with a linear amplifier and 
the kinescope transfer characteristics, 
EJE = 0.05, of Fig. 13 furnish the 
curve family shown in Fig. 16. The 
parameter in this curve is the high-light 
exposure in the camera tube with re- 
spect to its "knee point." The values 
1.2 to 7.2 represent a range of five lens 
stops. The optimum exposure for best 
tone quality and texture is near the 
value ffj/BkM. = 2 (see Figs. 9a and 
9b). Overexposures, 4.7 and higher, 
result in excessive electron flare (redis- 
tribution) and poor quality (see Figs. 
8a and 8b); underexposures, 1 or less, 
result in loss of shadow detail. 

* The exposure, E, is given in meter-candle 
seconds. 



Otto H. Schade: Transfer Characteristics 



153 



0.01 



CAMERA TUBE: IMAGE ORTHICON 
AMPLIFIER: LINEAR (fr=l) 
KINESCOPE 



12 



4.7 




z 



0.01 



Fig. 16. Over-all 
transfer charac- 
teristics of a tele- 
vision system. 



RELATIVE EXPOSURE (E,/EJ) 



Adjustment of the black-level setting, 
E , at the kinescope changes its transfer 
characteristic (see Fig. 13) and, conse- 
quently, the gamma of the over-all 
transfer characteristic. The charac- 
teristics obtained with linear amplifiers 
and with the kinescope black-level 
setting, E /E, as parameter are shown in 
Fig. 17 by the broken-line curves for an 
image signal source with linear transfer 
characteristic, yi = 1, such as a light- 
spot scanner or orthicon (or British 
C.P.S. Emitron), and by the solid 
curves for normal exposure of an image 
orthicon. The characteristics for the 
linear signal source are identical with 
the kinescope operating characteristic 
and, because of the high gamma, y lz ^ 
2.2, seriously compress tone rendition of 
scenes exceeding a 10 to 1 contrast range 
(blocked "blacks")- The image orthi- 
con characteristics are much more 
acceptable because they permit repro- 
duction of a scene having a contrast 
range of 40 to 1 with an over-all gamma 
of approximately 1.3 decreasing in the 
highlight and shadow tones. A natural- 
tone rendition (constant gamma) re- 
quires, therefore, a correction of the 



transfer characteristic. More specifi- 
cally, reproduction of a scene having a 
contrast range of 100 to 1 and constant 
gamma requires an over-all gamma of 
unity. 

The characteristic of a standard 
motion picture film process is shown in 
Fig. 18. The film characteristic, 0, is 
slightly ^-shaped and the over-all sys- 
tem characteristic, curve 1 or 2, be- 
comes more /S-shaped because of the 
flare light bias from camera- and projec- 
tion-lenses ( l /2% for each) and the light 
bias due to ambient light on the projec- 
tion screen ( 1 %) . A combination and a 
repetition of unconnected television and 
motion picture processes result in a 
more serious compression of both shadow 
and highlight gradation as exemplified 
by the combination characteristics in 
Fig. 19. 

It is evident that a normal projected 
35-mm motion picture is not an ideal 
source for generating television signals 
and neither is it, as is well known, a 
good source for making a duplicate 
motion picture. It is common knowl- 
edge that a chain of separate amplifiers 
can remain linear with respect to trans- 



154 



February 1951 Journal of the SMPTE Vol. 56 



<CD 



O.I 



UJ 

> 6 



0.01 



CURVE 



Knee 



2.3 



SIGNAL SOURCE KINESCOPE (<>/) 



LINEAR (ORTHICON OR 
LIGHT -SPOT SCANNER 



IMAGE ORTHICON 



d*0.05 

b=0.09 



a = o.os 

b 0.09 



o.i3 

0.30 







0.01 



4 66 




E |0 /E,=0.05 



O.I 



6 a 



1.0 



RELATIVE EXPOSURE (,/,) 



Fig. 17. Over-all transfer characteristics of television systems 
with image orthicon and linear signal sources. 



fer and frequency characteristics when 
each amplifier has linear response 
characteristics. Such amplifiers can be 
cascaded without loss of quality. By 
the same principle, imaging processes 
with linear response characteristics can 
be cascaded without loss of quality. It 
is possible to correct errors in the fre- 
quency response and transfer charac- 
teristics, but the image signals must re- 
main considerably above the fluctuation 
or "noise" level at all points of the sys- 
tem after passing through lenses, films, 
television tubes and amplifiers. 



6. Transfer Characteristics and Gam- 
ma of Motion Picture Film for TV 

The reproduction of images over a 
motion picture and television process 
involves a large number of transfer ele- 
ments. Shape and contrast range of 
the transfer characteristic of a normal 
motion picture positive are adjusted to 
fit the optical conditions in direct 
theater projection. It is logical, there- 
fore, that the characteristics of motion 
picture film intended as a picture 
source for reproduction by a television 
system or for storage and reproduction 



Otto H. Schade: Transfer Characteristics 



155 




(SS300dd 1SVT) 



iZd 







^i ^tt 

1 I 



s II 



< 



5 

SS3NlHOIfc)8 



3AIXV13U 



156 



February 1951 Journal of the SMPTE Vol. 56 



of video signals should be adjusted to fit 
the range and transfer characteristic 
of the television system and not the 
eye. The addition of one or several 
imaging processes increases the need for 
low distortion. The television process 
introduces as an important parameter 
an adjustable "black level" which, in 
effect, permits a subtraction of light 
levels and eliminates high densities in 
the positive film as a requirement. 
The "signal" levels in the system (con- 
trast range) can be reduced to lower 
values to gain linearity of signal transfer 
as in electron tube amplifiers, but the 
signal must remain sufficiently large to 
prevent an increase of the fluctuation 
level (noise, treated in Part II). 

The excessive distortion of the tone 
scale (see Fig. 19) resulting from an 
addition or a repetition of "normal" 
processes can be prevented by restricting 
the operating range on tube and film 
characteristics. With regard to the film 
process it is evident that operation in 
the constant-gamma sections of the film 
transfer characteristics results in uncrit- 
ical exposure conditions and in over-all 
film characteristics with unity or con- 
stant gamma. Inspection of available 
film characteristics shows that the con- 
stant-gamma range of positive films in 
particular extends over hardly more 
than a 20 to 1 range in transmission. 
It is very desirable that films with a 



shorter toe and a longer constant-gamma 
range be developed for television pur- 
poses. 

7. Motion Picture Film for Television 

The graphic solution for the optimum 
density range and gamma of motion 
picture positive film for television is 
quite simple. Because it is advan- 
tageous to use a substantially linear 
amplifier, the density range of the posi- 
tive film (controlled by the print 
gamma) should be adjusted to equal the 
optimum exposure range of the camera 
tube which can be determined from the 
over-all transfer characteristics such as 
those shown in Fig. 17. The exposure 
scale is adjusted by adding or sub- 
tracting a constant, EIQ, to the exposure 
values. The constant is selected to ob- 
tain a characteristic with reasonably 
constant gamma. An image orthicon 
(curve 2.3a, Fig. 17) requires an addi- 
tive constant, E w /Ei = 0.03, i.e., an 
exposure contraction to a 25 to 1 range, 
while an orthicon type (Fig. 17), will 
give a more constant gamma with E\Q/ 

>\ 

EI 0.05 and an exposure range of 
approximately 10 to 1. The density 
range, AZ) 2 , in the positive film should 
thus have the values AZ) 2 1.4 for an 
image orthicon and AZ) 2 = 1 for a linear 
camera tube such as the orthicon. De- 
sirable motion picture film character- 
istics are shown in Table II. 



Table II. Desirable Characteristics of Motion Picture Film for Television. 

Camera Tube 





Iconoscope 


Image 
Orthicon 


Orthicon 


Remarks 


Camera exposure range 


30 to 1 


25 to 1 


10 to 1 


Linear amplifier 


Positive Film 








Constant gamma, 
short toe 



Density range AD 2 



1.6 



1.4 



Between shoulder 
and toe 



Approximate gamma (72) 
for negative ADj = 1.26 

for negative ADi = 0.95 


1.28 
1.7 


1.12 
1.47 


0.8 
1.05 


for 71 = 0.68 
Exposure range 
of neg. 100:1 
30:1 



Otto H. Schade: Transfer Characteristics 



157 



Fig. 20 is on plate page 169. 




0.01 



Fig. 21. Operating 
characteristics of kine- 
scope with restricted 
signal range. 



0.01 



RELATIVE VIDEO SIGNAL VOLTS (E/E) 



It is obvious that the specified density 
range, AD 2 , in the positive print requires 
adjustment of the print gamma, 72, 
which depends on the density range of 
the negative which, in turn, is a function 
of negative exposure and brightness 
range in the camera image (100 to 1 and 
30 to 1 for 71 = 0.68 in Table II). A 
35-mm print of the SMPTE Television 
Test Negative on 5365 stock developed 
to a gamma of 0.95 gave excellent uni- 
formity of levels and negligible "flare" 
when projected into an orthicon or image 
orthicon. The reproductions of the gray 
scale and tone values in the picture sec- 
tion were excellent (Fig. 20),* required 
no black expansion, and were far supe- 
rior to those obtained with a normal high- 
gamma test film. Due to absence of 
distortion in the constant-gamma posi- 
tive, however, black-level and signal 
range varied in accordance with vari- 
ations hi density and range in the nega- 

* The arc line in the picture is a target 
flow in the camera tube. 



tive, exceeding at times the exposure 
range of the orthicon. Reproduction 
of the film by an image orthicon with 
moderate exposure (Ei/E*m = 1.2) 
gave, therefore, the best results. The 
appearance of an optical projection of 
the low-gamma print is, of course, 
quite unsatisfactory. 

The process of kinescope recording is, 
in principle, a process for storing and 
duplicating a video signal. Assuming a 
constant-gamma transfer of light values 
by the film process, it follows that the 
combination transfer characteristic of 
the recording kinescope, the film-scan- 
ning camera tube, and the video ampli- 
fier must likewise have a constant 
gamma to obtain an undistorted dupli- 
cation of the original video signal. It 
has been shown that the adjustment of 
the operating point, E , on the kinescope 
characteristic is a means of varying con- 
trast range and gamma of the kinescope 
and results in a family of characteristics 
shown in Fig. 21. The exposure range 



158 



February 1951 Journal of the SMPTE Vol. 56 



Fig. 22. Operating charac- 
teristics of image orthicon 
with restricted exposure 
range. 

of the film and the following camera tube 
can thus be restricted to any desired 
value. 

The light level representing zero video 
signal at the kinescope is a gray level on 
the positive film and results in a (d-c) 
signal-current level on the camera-tube 
characteristic below which the video 
signal will never decrease. This mini- 
mum signal level, I , represents, there- 
fore, the black level in the video signal, 
and is set to zero at the transmitter by 
an operator adjustment. Restriction 
of the exposure range and resetting of 
the black level by subtraction of the 
minimum signal, I , from the transfer 
characteristic of the camera tube fur- 
nishes the family of operating charac- 
teristics as shown in Fig. 12 or as in 
Fig. 22 for an image orthicon. The 
method of constructing these character- 
istics has been explained above (see 
Eq. 12a). 

After the curve families are plotted, 
the light range giving the best match of 
kinescope and camera-tube operating 
characteristics can be determined by 
rotating one of the curve sheets 180 
around its diagonal and placing it over 
the other curves so that the scales for 



L IMAGE ORTHICON 

TARGET BIAS 2 VOLTS ABOVE 
CUTOFF 




RELATIVE EXPOSURE (E,/E,) 



light values and electrical signal scales 
superimpose. For the example, the 
curves coinciding most accurately in 
shape are the kinescope curve, E /E = 
0.3, and the camera-tube curve, IJ1 = 
0.1. (It is permissible to twist* the 
coordinates slightly.) When equal video 
signal values are selected, the corre- 
sponding light values on the two char- 
acteristics plotted against each other fur- 
nish the required transfer character- 
istic for the motion picture process (Fig. 
23), assuming that a linear video ampli- 
fier is used. Film characteristics ap- 
proaching this characteristic can now be 
selected. For uncritical exposure con- 
ditions and processing, both negative 
and positive films should be developed 
to approximately equal and constant 
gammas. It will be shown in Part II 
that a higher negative gamma and short 
toes reduce the fluctuation (noise) level 
in the film process. The remaining error 
in the film transfer characteristics is to 
be eliminated by correcting the transfer 
characteristic of the video amplifier 

* The angle of twist indicates the depar- 
ture from unity gamma in the associated 
photographic process. 



Otto H. Schade: Transfer Characteristics 



159 



5 



^Z 
uO 
ZO 

II 

s 



ofe 

"8 
ijl 

is 



5- 
& 



2 



CURVE 


KINESCOPE o/E) 


IMAGE ORTHICON (^/l) 







0.3 


ai 







0.25 


0.05 




VIDEO SIGNAL RANGE 100: 


LINEAR AMPLIFIER 




















^/ 


















/ 


















< 


? 
















/ 


' 


















/ 


/ 




















/ 


















4 


/ 


















/ 
















y 


/ 




































2 


4 66 


2 468 



0.01 



01 

RELATIVE EXPOSURE OF NEGATIVE 



1.0 



Fig. 23. Over-all transfer 
characteristic of motion 
picture process for repro- 
duction of television re- 
cording with image or t hi - 
con. 



preceding the kinescope or following 
the camera tube or both. Depending 
on the position in the chain (video sig- 
nal-amplifier-kinescope-film process- 
camera tube-amplifier-video signal), 
the video correction characteristics 
differ. Correction of the amplifier 
driving the recording kinescope, for 
example, assumes that the video signal 
output values on the camera-tube char- 
acteristic and the video-input signals to 
the kinescope amplifier are equal. The 
amplifier output signal is thus found by 
tracing the camera signal over the light 
values in the film and kinescope char- 
acteristics back to the kinescope grid 
signal which equals the amplifier out- 
put signal. An electrical correction 
following the film process is more practi- 
cal and has the advantage that the final 
image is under direct visual observation, 
permitting instantaneous adjustments 
for best results. 

When an iconoscope is used as a film 
scanner the best compromise match of 
characteristics is obtained for EJE ~ 



0.3 at the kinescope and an averaged 
peak illumination, Ei, in the order of 
10 units (in Fig. 11) at the iconoscope 
mosaic. The gamma in the high-light 
range of the iconoscope, however, is 
too low, requiring considerable "white" 
expansion by a correction amplifier. 
This condition is amplified at higher 
peak exposures of the tube for which 
the kinescope bias decreases toward 
EJE = 0.2. The operating conditions 
for a linear film scanner (orthicon, light- 
spot scanner) are evaluated similarly 
and furnish the transfer characteristics 
shown in Figs. 24a and 24b. The film 
characteristics are listed in Table III. 
Film characteristics for a camera tube 
with constant gamma, 71 = 0.5, are 
listed for comparison. 

8. Effect of the Line Raster on Film 
Exposure and Sensitometry 

Maximum detail contrast in kinescope 
images requires a scanning-beam diam- 
eter smaller than the pitch distance of 
the scanning or "raster" lines. (See 



160 



February 1951 Journal of the SMPTE Vol. 56 



Fig. 24a. Over-all trans- 
fer characteristic of 
motion picture process 
for reproduction of 
television recording 
with linear camera 
tube. 



RELATIVE TRANSMITTANCE (T,) 
OF POSITIVE 

O ~ 

_ M Jk o> O 


\ 


CURVE 


KINESCOPE (E /) 


IMAGE ORTHICON (lo/ l] 




ETL 


0.3 
0.3 
0.25 


0. 1 
0.05 
0.05 


'IDEO SIGNAL RANGE 100:1 


















f 


s 
















<s 


















^ 


















f? 














/ 


f 


r 
f 
















/ 


t 














- 


/ 


/ 


















2 


4 e e 


2 4 e 



0.01 OJ 

RELATIVE EXPOSURE (,/,") 
OF NEGATIVE 



1.0 



Fig. 24b. Transfer char- 
acteristics of film proc- 
ess and amplifier used 
in process of Fig. 24a. 



RELATIVE AMPLIFIER OUTPUT VOLTS C 
RELATIVE FILM TRANSMITTANCE (T 2 ) 

o P M * 6 




CURVE 


KINE SCOPE (E O /E) 


IMAGE ORTHICON (I /l) 


^^ 


0.3 


O.I 


VIDEO SIGNAL RANGE 100-1 


712 = 0.7 


















/ 

s 
















^ 


S 














*4 

/r 


s/ 

& 














A 










X 


x 


X 


/ 

/ 
/ 








01 


z 


4 ' 8 o. 


4 -. e 



RELATIVE AMPLIFIER INPUT VOLTS OR 
RELATIVE FILM EXPOSURE (E,/E,) 



Table III. Characteristics of Motion Picture Film For Video Recording. 





Camera 
Tube 
71 = 0.5 


Iconoscope 


Image 
Orthicon 


Orthicon 
71 = 1.0 


Remarks 


Exposure range in 
negative 
Negative 71 
Positive 72 
Over-all 712 
Density range in 
positive 


50:1 
0.875 
1.25 
1.1 

1.6 


30:1 
1.0 

1.25 
1.25* 

1.85 


20:1 
1.0 
1.1-1.2 
1.1-1.21 

1.4 


20:1 
1.0 
0.7 
0.7J 

0.9 


Constant 
Constant 
Constant 



* Requires electrical expansion in the high-light range, 
t See Fig. 23. 
J See Fig. 24. 



Otto H. Schade: Transfer Characteristics 



161 



Reference 1 and Part IlFof this paper.) 
A kinescope beam defocused or verti- 
cally enlarged jto produce a "flat" field 
has an equivalent rectangular cross 
section equal to the pitch distance of the 
raster lines. With a flat field a given 
brightness range in the kinescope image 
can be recorded on film with normal 
exposures. Normal sensitometric con- 
ditions are maintained also for the case 
in which a smaller scanning aperture 
produces a line structure on the kine- 
scope which is not resolved by the com- 
bination of lens and negative film, re- 
sulting again in a flat field on the nega- 
tive.* 

An equivalent line cross section, 
smaller than the raster pitch distance, 
requires an increase of the kinescope 
line brightness by the ratio of pitch dis- 
tance to line width in order to maintain 
a given image brightness. Hence, when 
a high-resolution kinescope is focused 
sharply and the dark spaces between 
raster lines are, for example, equal to the 
line width, the line intensity doubles. 
To record the increased line-brightness 
values within the normal range on the 
film transfer characteristic, the exposure 
of negative film with adequate resolu- 
tion (for example on a 4 X 5 in. film) 
must be reduced to one-half the value 
found normal for a flat field (defocused 
case). The negative is given a normal 
development but will appear underex- 
posed, because part of the film area re- 
mained unexposed by the black raster 
spaces. For a perfectly sharp recording, 
the minimum transmittance (r m in) of 
the negative cannot be expected to de- 
crease below the theoretical value : 
Tmin = (pitch distance) 

(equivalent line width) 

Measurement of true line-density values, 

* It is noted that a "flat" field requires a 
constant spot diameter and perfect uni- 
formity of line spacing, conditions which 
are difficult to realize with rasters con- 
taining 500 or more lines. The effects 
on resolution will be discussed in Part 
III. 



D or D m ax, requires use of a microdensi- 
tometer to read an area within the line 
cross section. 

Printing of a sharp line raster nega- 
tive with normal line densities on a posi- 
tive film requires a normal exposure, 
but the "weighted" transmittance of the 
positive cannot exceed the maximum 
value : 
Tma* = (pitch distance) 

(equivalent line width) 
irrespective of the resolution of this 
process. Projection of the positive re- 
quires a higher film illumination, but 
contrast range and over-all transfer 
characteristic are normal. A print on 
paper, however, results in a reduced con- 
trast range because of the unchanged 
"black" limit of the paper. High 
lights are "gray" due to light trans- 
mission and exposure through the clear 
spaces separating the exposed raster 
lines in the negative. In practice, 
neither kinescopes, lenses nor negative 
film can maintain perfect contrast 
between raster lines and "black" spaces 
throughout the tone range. Practical 
"resolving apertures" (see Section A) 
have a nonuniform light distribution 
which causes a gradual change of light 
intensity and exposure between line 
centers. Test exposures on 4 X 5 Super 
XX Film made with a defocused kine- 
scope spot and, subsequently, with a 
sharply focused spot approximately 
^j line-pitch in diameter, have shown 
that an exposure reduction by a factor of 
two for the sharp-line negative resulted 
in a slightly higher density of the scan- 
ning lines and a "weighted" density 
range of 0.19 to 1.0; the density range 
of the defocused negative was 0.3 to 1.2. 
The two negatives received identical 
processing. Contact positives were 
made by printing both negatives side 
by side on one sheet of film. The 
weighted densities, D, of the fine-line 
positive were higher by AD c^. 0.3 (due 
to the black spaces between lines). 
With correspondingly adjusted illumina- 
tion, the tone scales of the two positives 



162 



February 1951 Journal of the SMPTE Vol. 56 



Fig. 25. Transfer char- 
acteristics of gamma- 
correction amplifier 
providing over-all 
gamma of unity. 





CURVE 


SIGNAL SOURCE 






1.0 



(0 

o 
0.02 H 

0.012 < 





LINEAR 







IMAGE ORTHICOr 


J 


KINESCOPE Eo/E=0.05 




















^ 


^ 
















x 


^ 


/^ 
















^ 


X 


^ 












X 


^ X 

5 
t 


X 

\^ 


^ 














X 


y> 


^ 


^ 














/ 


/ 


'/ 


















/ 


/ / 




















/ 


/ / 


















6 

4 

2 

0.01 


/ 


// 


















bl 
CC 


1 
1 








































2 


4 e 


2 468 





0.01 



O.I 
RELATIVE VIDEO SIGNAL INPUT 



1.0 



were substantially the same at a normal 
viewing distance, the fine-line positive 
being preferable for its sharpness. 
The effective transfer characteristic of an 
imaging process transducing a raster 
image with a nonuniform intensity dis- 
tribution over the line cross section is 
actually a weighted transfer character- 
istic extending over a greater range of 
signal values. The practical equivalent 
for particular cases is best determined 
by measurement. 

Experience with photography of sharp 
kinescope images substantiates the re- 
quirement of a transfer characteristic 
with a somewhat longer range or, as 
pointed out above, a reduction of the 
kinescope brightness range (raised black 
level). 

9. Gamma Correction in Television 
Systems 

The point gamma values in the over- 
all transfer characteristic are the prod- 
uct of the respective point gammas of all 
component characteristics. A low 



gamma value in one element can, there- 
fore, be corrected by a high (reciprocal) 
gamma value in another element. A 
distorted over-all response character- 
istic can be made linear by a response 
characteristic having reciprocal point 
gamma values. Gamma correction char- 
acteristics for several normal television 
system characteristics are shown in Fig. 
25. The characteristics are constructed 
by selecting equal light values for input 
and output of the system and plotting 
the corresponding camera-tube output 
signal (input to the amplifier) versus the 
input signal (output of the amplifier) of 
the kinescope. (A scale indicates cor- 
responding optical signal values.) 

The transfer characteristics of a video 
amplifier can, in principle, be modified 
by the use of nonlinear circuit elements 
to have any desired shape. Curved 
transfer characteristics can be obtained 
by utilizing the normal curvature of one 
or several electron tubes in parallel or 
in series. Greater curvatures can be 
produced by nonlinear resistances such 



Otto H. Schade: Transfer Characteristics 



163 



LEVEL SETTER 
OR CLAMP CIRCUIT 




EC MAX 



Fig. 26. Gamma-correction circuit 
providing for black-and-white 
expansion in a single stage. 

as diodes or triodes in combination with 
resistors, employed as load resistances 
in current-carrying electrodes of elec- 
tron tubes. A circuit permitting the 
correction of single curvature as well as 
$-shape characteristics hi one ampli- 
fier stage is shown in Fig. 26. This cir- 
cuit has a relatively low insertion loss 
because the nonlinear elements are lo- 
cated hi the cathode circuit of the am- 
plifier permitting the use of compensated 
high-impedance plate loads which must 
remain constant. The voltage gain of 
such a correction stage for a linear sig- 
nal source is in the order of 0.75 for 
a 20-mc channel* and proportionally 
higher for lower passbands. The larger 
signal voltages required for correcting 
strong curvatures are handled easily by 
a small amplifier tube (6AC7). 

At zero signal input, diode 1 (black 
expander) of Fig. 26 is made to conduct 
a current the value of which is deter- 
mined by the voltage, E Dv The diode 
impedance shunts the cathode resistor, 
R k , with a relatively low value; the 
amplifier gain is high. When the ampli- 
tude of the signal current equals the 
initial diode current, diode 1 becomes 
an open circuit and the amplifier gain 

* This type of circuit has been in use by 
the author for many years. 



is degenerated to a lower level deter- 
mined by R k . At higher signal ampli- 
tudes the cathode potential rises and 
finally diode 2 (white expander) con- 
ducts, shunting R k by its impedance 
(adjustable by R t ), and thus increases 
again the amplifier gam. The graphic 
construction of the transfer character- 
istic from the conductance character- 
istics of the cathode circuit elements is 
illustrated by Fig. 27 for the case hi 
which the amplifier stage also performs 
the function of a black-level clipper. 
The current-voltage characteristic, l k vs. 
E k , of the cathode circuit results from 
simple addition of the currents of the 
parallel elements (vertical addition of 
curves R k , Di and Z> 2 ) ; the series char- 
acteristic, /* vs. E t , for tube and cathode 
circuit results from (horizontal) addi- 
tion of the grid-to-cathode voltage, 
E OK , of the tube and the cathode-to- 
ground voltage, E k , of the circuit. The 
transfer characteristic, I h vs. E., of plate 
current versus signal input voltage, E t , is 
finally obtained by subtracting the 
screen-grid current from corresponding 
cathode-current values. 

The graphic determination of the cir- 
cuit constants required for a given 
gamma correction (broken-line curve, 
Fig. 25) is illustrated in Figs. 28a, 28b 



164 



February 1951 Journal of the SMPTE Vol. 56 




-4 -3 -2 -I 

CONTROL-GRID VOLTS 



25 
20 
515 

H 

10 


5 

o 








I d| =5MA 


ED, = 5.5V 

R K =900 OHMS 
R s = 500 OHMS 












* 


E C =+I.3V 












*/ 




/ 


c 






/ 
/ 












7 


/ 


/ 








/ 
/ 










/ 





C ^' 


f^ 






/ 




A .< 


,.^' X 


B 


^ 


^^ 








1 




^~ 




x^ 




^^ 














4 


^ 


^ 


' I 


^ 


' 








, 






t 


t '// 


^ 


^ ^^ 


f^ 


DIOC 


)E \" 




s 


^D 


IODE 


2+500 
OHMS 




'.'' 


ft" 










/ 













VOLTS 



10 



Fig. 27. Graphic construction of transfer characteristic 
of gamma-correction circuit (Fig. 26). 



and 28c. Plate and cathode current for 
the amplifier tube (6AC7) are drawn in 
linear coordinates. When a zero-signal 
plate-current value (h = 6 ma for the 
example) is selected, the desired plate 
current versus signal characteristic, 
I b vs. E s , is drawn with a current range 
(6 to 16 ma) within the range of the 
tube. The voltage range is selected to 
avoid a crossover with the tube char- 
acteristic, I b , as shown in Fig. 28a.* 
The characteristic, I k vs. E t , is now con- 
structed by adding the screen-current 
values, 7 C2 , corresponding to the plate 
currents, I b . Subtraction of the grid- 
cathode voltage, E OK , (horizontally) 
from this curve results in the cathode 
network characteristic, I k vs. E k . A line 
drawn tangent to this curve furnishes 
the value of the cathode resistance load, 
R k (1450 ohms). The load, R k , inter- 
sects the current axis at E t = and a 
current value, /R A , equal to the total 
current in R k at zero signal (13.5 ma). 
The voltage, E k , between cathode and 
ground follows from Ohm's law (E k = 



* A lower voltage range will be found to 
require an amplifier tube having a steeper 
plate-current characteristic than shown. 



1450 X 0.0135 = +19.6 v). The dif- 
ference in current, In k I k , as a func- 
tion of E k is plotted by subtracting the 
curve, I k vs. E k , from the R k character- 
istic. It represents the required diode 
circuit characteristic, I d vs. E d (Fig. 
28b) . This characteristic is now broken 
up into sections which can be obtained 
with available diodes. Figure 28b shows 
the component curves Di, D 2 , Da, which 
can be obtained with germanium diodes 
and series resistances (Fig. 28c). The 
zero-signal diode currents (1.9, 1.0 and 
3.4 ma) are obtained from the curves 
as well as the diode biasing voltages 
which exceed E k (19.6 v) by the volt- 
age drop of the diode characteristics 
DI, Dz, DS* For the example, the diode 
bias potentials are: E Dl = +20.2 v, 
E D2 = +21.1 v, and E D3 = 23 v. 

The curve, I d vs. E d , can be approxi- 
mated with two diodes, but it is cau- 
tioned that too sharp a break in a trans- 
fer characteristic may cause a "quan- 
tizing" effect, which is a spurious con- 
tour at gradation values corresponding 
to the break point. 

The diagram of a practical gamma- 
correction circuit with adjustable black- 
and-white expansion is shown in Fig. 



Otto H. Schade: Transfer Characteristics 



165 





19.6 V 



'1450 



2 3 

DIODE VOLTS (Ed) 



1200 



1400 



Figure 28b. 



+ 211 V 



+ 20.2 V 
Figure 28c. 



+23 V 



Fig. 28. Graphic determination of circuit constants required 
for a given gamma correction. 



29. It is essential to maintain a con- 
stant operating point, Ib , on the tube 
characteristic by a level setter which 
establishes a fixed operating bias for 
the tube. The bias voltages for the 
black-expansion diodes, D\, Z) 2 , D 3 , are 



obtained from a tapped bleeder circuit 
in which a heavy bleeder current main- 
tains substantially constant potentials 
with varying diode currents. The 
black-expansion control changes the 
diode bias values in proportion, main- 



166 



February 1951 Journal of the SMPTE Vol. 56 



SIGNAL 
OUTPUT 




ALL RESISTOR 
VALUES IN OHMS 



Fig. 29. Gamma-correction circuit with adjustable black-and-white expansion. 



Fig. 30. Oscillograms of transfer Fig. 31. Oscillograms of transfer 
characteristics for various settings of characteristics for various settings 
black expansion control in Fig. 29. of curvature control in Fig. 29. 



Fig. 32. Oscillograms of transfer 
characteristics with black-and-white 
expansion obtained with Fig. 29. 




Otto H. Schade: Transfer Characteristics 



167 



taining a smooth curvature of the cor- 
rection characteristic as shown by 
the oscillograms given in Fig. 30 (ob- 
tained with a linear sawtooth-voltage 
input to a similar correction stage). 



Variation of the 250-ohm curvature 
control affects the relative diode po- 
tentials and, hence, the knee curvature 
of the characteristic (see Fig. 31). The 
correction circuit for expanding the high- 




WHITE LEVEL 



BLACK LEVEL 



I 
STEP TABLET DENSITY SCALE 



Fig. 33. Transfer curves of camera tube and video amplifier in semilog 
coordinates, showing an uncorrected characteristic, 0, and characteristic, C, 

with proper correction. 






168 



Fig. 34. Oscillograms of transfer characteristics of light-spot 

scanner (linear) and video amplifier with correction, C, 
and without correction, 0. Step tablet density range AD = 2. 

February 1951 Journal of the SMPTE Vol. 56 



light range (reversed diodes Z> 4 and D 5 ) 
is similarly adjustable. The setting of 
the white expansion control determines 
the point at which the gamma starts 
to increase and the slope control deter- 
mines the desired increase of the point 
gamma. The high-light diode circuit 
operates normally with small average 
currents and does not require a heavy 
bleeder current to obtain stable poten- 
tials. The oscillograms given in Fig. 
32 illustrate characteristics with black- 
and-white expansion. 

A practical method for adjusting 
gamma-correction stages is the observa- 
tion with an oscilloscope of the signal 
from a logarithmic step tablet. A 
linear transfer of signals between the 
point of observation and the kinescope 
control grid and a linear over-all trans- 
fer characteristic require a characteristic 
of voltage versus light input which is 
the inverse of the kinescope character- 



istic. When the light input is changed 
logarithmically along a cross section of 
the picture, the linear oscilloscope time 
base represents a log-scale. The oscillo- 
gram of the step tablet is, hence, a semi- 
log plot of the corrected camera transfer 
characteristic which, for unity over-all 
gamma, must be the inverse of the 
kinescope characteristic regardless of 
the type of camera tube or signal source 
used. The gamma-correction stage is, 
hence, adjusted to obtain the transfer 
curve, C, shown in Fig. 33. (Note the 
upward curvature of the steps.) The 
oscillogram obtained with linear ampli- 
fiers from a linear signal source (light- 
spot scanner or orthicon) is shown for 
comparison. Actual oscillograms for 
these cases are given in Fig. 34. 

The effectiveness of gamma correction 
in the amplifier is demonstrated by 
comparing the photographs given in 
Figs. 35 and 36. The photographs, 




Fig. 20. Television reproduction of 35- mm motion picture frame 

from special low-gamma SMPTE test film. Bandwidth = 10 megacycles; 

scanning lines = 525; interlaced 2:1. 



Otto H. Schade: Transfer Characteristics 



169 





170 



Fig. 8a, b. Poor quality television picture caused by over- 
exposure of image orthicon. Bandwidth = 4.25 megacycles; 
scanning lines = 525; interlaced 2:1. 

February 1951 Journal of the SMPTE Vol. 56 





Fig. 9a, b. Good quality television picture obtained with proper 

exposure of image orthicon. Bandwidth = 4.25 megacycles; 

scanning lines = 525; interlaced 2:1. 

Otto H. Schade: Transfer Characteristics 



171 





Fig. 35a, b. Photograph of optically projected 2 X 2 in. slide. 
172 February 1951 Journal of the SMPTE Vol. 56 





Fig. 36a, b. Photograph of television reproduction of 2 X 2 in. slide; 

light-spot scanner with gamma correction. Bandwidth = 4.25 megacycles; 

scanning lines = 525; interlaced 2:1. 



Otto H. Schade: Transfer Characteristics 



173 



CAMERA 



MONITOR KINESCOPE 




HOR SYNC- 



Fig. 37. Block diagram of vertical-cross-section selector 
and signal -measuring system. 




Fig. 38. Step tablet and calibration pulse on kinescope screen. 
174 February 1951 Journal of the SMPTE Vol. 56 



Figs. 35a and 35b, were made of a direct 
optical projection (coated lens, Koda- 
slide Projector Model 24) of two excel- 
lent miniature slides (obtained from the 
Eastman Kodak Co.) on a matte paper 
screen. The television reproductions, 
Figs. 36a and 36b, made with a corrected 
light-spot scanner of the same slides, are 
photographs of a 16-in., 525-line kine- 
scope image reproduced over a stand- 
ard 4.25-mc television channel. A nega- 
tive film size of 4 X 5 in. and a 1.5-sec 
time exposure at //16 minimize deterio- 



ration of image sharpness by the photo- 
graphic process which was identical in 
both cases. 

It is cautioned that the correction of 
transfer characteristics is accompanied 
by a change of the "noise "-to-signal 
ratio proportional to the gamma (7) of 
the correction characteristic. The sub- 
jects of random fluctuations in the tele- 
vision and photographic processes as 
well as resolution and detail contrast 
will be discussed in Parts II and III of 
this paper. 



APPENDIX 



1. Measurement of Camera-Tube 
Transfer Characteristics 

An illuminated step tablet covering a 
range of 100 to 1 in ten equal logarithmic 
steps (AD = 0.2) is placed vertically 
in the viewing field of the camera. The 
magnification is adjusted for a step size 
of 5 to 10% of the picture width on the 
kinescope. A vertical cross section of 
the step signal is obtained on an oscillo- 
scope operated with a 60-cycle linear 
time base by applying to the oscilloscope 
control grid a positive horizontal pulse 
voltage of 5 to 10% of the line duration, 
timed by a double line frequency pulse 
to occur in the center of the field. The 
strip image from the camera is displaced 
horizontally against the vertical cross 
section selected by the pulse so that part 
of a uniform background adjacent to 
the step tablet produces a reference 
level signal on the oscilloscope (see Fig. 
34). Depending on the background 



along the step tablet, the reference sig- 
nal represents a black, white or gray 
level against which the step signals 
are measured. To eliminate errors due 
to amplifier nonlinearity and to permit 
expansion of small signals, the step 
signals are measured by a substitution 
method. A "notched" calibration pulse, 
several scanning lines wide, repeating at 
60 cycles/sec is superimposed on the 
low-level signal from the camera (Fig. 
37). It appears on the kinescope as a 
horizontal strip (see Fig. 38) which can 
be made to pass over any step of the 
tablet by vertical displacement of the 
tablet or its image (panning). The 
calibration pulse displaces a section of 
the oscillograph trace vertically (Fig. 
39) permitting it to "lift" or depress 
(when of opposite polarity) a section of 
the reference level into the step level. 
The signal difference is thus equal to 
the pulse voltage. A convenient method 





a. Partial displacement. b. Pulse amplitude equal to step signal. 

Fig. 39. Oscillogram of step tablet and superimposed calibration pulse. 

Otto H. Schade: Transfer Characteristics 175 



for reading the relative calibration pulse 
voltage is indicated in Fig. 37. Po- 
tentiometer P 2 (200 ohms) is a voltage 
divider for the pulse voltage and an 
auxiliary d-c voltage which is adjusted 
by the series resistance, R s , to give full- 
scale deflection on a d-c meter (0-100 
n&) when P 2 is set for maximum output. 
With the selector pulse superimposed 
on the highest step level from the refer- 
ence signal line, the maximum pulse 
signal lifting the reference level into the 
step level is set by potentiometer PI. 
All other levels are then matched by 
adjusting P 2 and the d-c meter reads 
directly the relative step intensities. 

Signal increments for adjacent steps 
can be measured by the same method. 
It must be considered, however, that 
redistribution, flare or edge effects in 
camera tubes can tilt the step signals.* 
The measurement of signals with re- 
spect to an adjacent level representing a 
constant light value at the source is, 
therefore, less likely to introduce errors 
because it is independent of shading. 
The characteristics of the image orthi- 
con (Fig. 6) were measured by this 
method. Target bias values of less 
than 2 v result in widely different char- 
acteristics for dark and light back- 
grounds (see references 1 and 3) when 
it is attempted to cover a large light 
range. 

2. Operation Requirements for Sys- 
tems With Gamma-Correction 
Amplifiers 

(a) Black-level variations are ampli- 
fied. The controlling operator must be 
able to observe the corrected video 
signal. 

(b) The signal gain between camera 
and correcting circuit should remain 
fixed once it is adjusted to give the cor- 
rect relation between transfer character- 
istics. Adjustment of signal ampli- 
tudes from camera tubes with non- 
linear characteristics should be made 
by controlling the exposure (mechanical 

* See Reference 1, Part IV, Fig. 81. 



or electronic shutter, iris, or neutral 
filters). 

(c) Special effects requiring resetting 
of exposure, gain, or correction as well 
as equalization of a bank of cameras, 
may require a separate quality-control 
position. In this case, the camera video 
man exercises control over electrical 
focus, beam current and shading. He 
should see the corrected picture and 
may control target bias according to 
instructions from the quality-control 
operator. 

(d) Expanded sections of the tone 
scale have a higher "noise" level, and 
compressed sections have a lower noise 
level. (See Part II.) A normal cor- 
rected tone scale results, in most cases, 
in a lower average brightness of the 
image than that obtained by overex- 
posure of the image orthicon, although 
the highlight brightness is held constant. 
There may be a tendency to increase the 
picture brightness by increasing the 
signal amplitude to the kinescope, which 
results in higher visibility of fluctuations 
(noise) . 

The British and other European 
television systems operating with a 50- 
cycle frame frequency are limited by 
flicker to an image brightness which 
is lower by a factor of six to ten in com- 
parison with American standards. A 
low picture brightness lowers the thresh- 
old of the eye for observing fine detail 
and fluctuations giving, therefore, the 
impression of a higher quality level. 

3. Calculation of the Average Number 
of Quanta in One Lumen of White 
Light 

The ratio of the luminous flux, F, to 
the radiated flux, P, within given wave- 
length limits, A\, is usually expressed in 
lumens per watt (Im/w) : 



= 680(F / F r ) ( AX)lm/w 

where YI = relative luminosity factor 

Y r = relative energy factor of 

source 

(FjF r )(AX) = average value of products in 
the range AX. 



176 



February 1951 Journal of the SMPTE Vol. 56 



The factors, Fj and F r , are unity at the 
reference wavelength, X = 0.555 /x. 
The radiant energy giving one 1m of light 
within the limits, AX, is hence : 



P(AX) = 0.00147/(FjF r )AX = 



1.47 X 10 4 /(F*F r )(AX) ergs/sec 

The energy per quantum is the product 
of Planck's constant: 

h = 6.547 X 10 ~ ergs/sec 

and the frequency of the light v = 3 X 
10 1 4 /X (X in microns) : 

hv = 19.641 X 10 ~ 13 A ergs/sec 

Expressed with respect to the reference 
wavelength X = 0.555: 

hv = 3.54 X 10 ~ 12 (0.555 A) 

where (0.555A) is the relative energy 
factor. 

The number of quanta in the radiated 
energy yielding one lumen of light is 
hence : 



= (1.47 X 



n r (AX) = 

10V3.54 X 

ttr(AX) = 4.16 X 

10 15 /((F, 



Y r 



/(AX) 



(AX) 



quanta/lm 



where the bracketed term is the average 
product of the relative energy factors 
over the range, AX. For noon sunlight 
or a black body at 5400 K and the 
limits X = 0.40 to 0.73 M, the above 
equation gives the quantum number: 

n,(AX) ^ 1.3 X 10 16 quanta/lm 

References 

1. O. H. Schade, "Electro-optical char- 
acteristics of television systems: Intro- 
duction," RCA Rev., vol. IX, no. 1, pp. 
5-13, Mar. 1948. 

"Part I Characteristics of vision 
and visual systems," ibid., pp. 13-37, 
Mar. 1948. 

"Part II Electro-optical specifica- 
tions for television systems," ibid., 
no. 2, pp. 245-286, June 1948. 
"Part III Electro-optical character- 
istics of camera systems," ibid., no. 
3, pp. 490-530, Sept. 1948. 
"Part IV Correlation and evaluation 
of electro-optical characteristics of 
imaging systems," ibid., no. 4, pp. 
653-686, Dec. 1948. 

2. P. K. Weimer, S. V. Forgue and R. R. 
Goodrich, "The Vidicon photoconduc- 
tive camera tube," Electronics, vol. 23, 
no. 5, pp. 70-73, May 1950. 

3. R. B. Janes and A. A. Rotow, "Light 
transfer characteristics of image orthi- 
cons," RCA Rev., vol. XI, no. 3, pp. 
364-376, Sept. 1950. 



Otto H. Schade: Transfer Characteristics 



177 



Technical Activities of the 
Motion Picture Research Council 

By W. F. Kelley and W. V. Wolfe 



A brief description is provided for the more important technical activities, 
including work in progress on composite photography, lighting equipment, 
wind machines, strippable adhesives, transportation equipment, plastics, 
diffusion cloths and other items of interest in the production of motion 
pictures. Mention is also made of the Research Council's function in 
connection with new products, inventions, television, stereoscopic motion 
pictures, standards and test films. 



ALTHOUGH the history and functions 
of the Motion Picture Research 
Council were covered in an earlier paper 
before this Society, 1 it seems desirable 
to review briefly that information. 

Originally, there was formed about 
1928 the Technical Bureau of the Acad- 
emy of Motion Picture Arts and 
Sciences. In 1932 the name, along 
with some of its functions, was changed, 
forming the Research Council of the 
Academy of Motion Picture Arts and 
Sciences. For fifteen years the Re- 
search Council existed largely as a secre- 
tariat relying upon voluntary services 
of studio and equipment manufac- 
turers' personnel for most of the work 
carried on. 

By 1947 it was freely recognized that 
insofar as methods, processes and equip- 
ment are concerned, there was no need 
for competition among the producers 



Presented on October 19, 1950, at the 
Society's Convention at Lake Placid, 
N. Y., by W. F. Kelley and W. V. Wolfe, 
Motion Picture Research Council, Inc., 
1421 N. Western Ave., Hollywood 27, 
Calif. 



of motion pictures. Accordingly, it 
was practical to carry on the develop- 
ment of such equipment, processes 
and methods in a common industry- 
sponsored technical organization. With 
this end in view, the Motion Picture 
Research Council was separated from 
the Academy of Motion Picture Arts 
and Sciences and incorporated under 
the laws of the State of California. 
Funds and facilities were made avail- 
able and the business of organizing a 
staff of qualified technical people and 
securing for them the necessary equip- 
ment and quarters was undertaken. 

The Research Council is interested in 
any and all technical problems in the 
production or exhibition of motion 
pictures. In general, the activities can 
be divided into three groups: service 
functions, short-range development and 
design problems, and long-range ad- 
vanced development problems. The 
staff includes two physicists, three chem- 
ists, two mechanical engineers, two elec- 
trical engineers and supporting person- 
nel. 

Figure 1 shows the large half of the 
mechanical and electrical laboratory 



178 



February 1951 Journal of the SMPTE Vol. 56 




Fig. 1. Mechanical and electrical laboratory. 




Fig. 2. Chemistry laboratory. 
Kelley and Wolfe: Research Council Activities 



179 



while Fig. 2 is one small corner of the 
well-equipped chemistry laboratory. 
Formalized laboratories of this type do 
not fit too well the type of work carried 
on by the Research Council since 
much of this work involves problems 
which can be properly studied only on a 
motion picture stage which, fortunately, 
is available through the cooperation of 
20th Century-Fox on whose Hollywood 
Western Avenue lot is located the Re- 
search Council's office. 

Although the Research Council now 
has its own technical staff and facili- 
ties, it needs the guidance of the many 
expert technicians of the industry. 
This is provided through a group of 14 
basic committees covering every phase 
of the technical activity of the industry. 

The Research Council is a small or- 
ganization covering a broad and diverse 
field. Its only possible chance of work- 
ing successfully under such conditions 
lies in the cooperation which it seeks 
and receives from other industries 
throughout the country. Since it is 
the purpose of the Research Council to 
serve the motion picture industry, it is 
not concerned with glory in solving 
problems, but only with the solution. 
If any other organization has a satis- 
factory answer, then the aims of the 
Research Council have been completely 
satisfied when that answer is made 
available to the industry. The cooper- 
ation of film manufacturers, equipment 
manufacturers, chemical companies and 
many others too numerous to name is 
gratefully acknowledged and deeply 
appreciated. 

Set Lighting 

Projects of many types and varieties 
are undertaken by the Research Coun- 
cil, either on its own or in cooperation 
with other companies. For example, 
set lighting is one of our most important 
projects. We will be concerned with it 
as long as there is a motion picture in- 
dustry. Presently we are carrying on 
work on set lighting in all three branches 



of our activity, that is to say, service 
function, short-range design and devel- 
opment and long-range advanced de- 
velopment. Some time ago, the in- 
dustry became seriously interested in 
the use of sealed-beam lamps and a 
very careful study of their application 
was made. This was reported in a 
paper before this Society at the Fall 
Convention in 1949. 2 Figure 3 shows 
an actual motion picture set at Para- 
mount Studio which was arranged to be 
lighted by either sealed-beam or stand- 
ard studio lamps in order that photo- 
graphic tests might be made under ac- 
tual operating conditions. Figure 4 
shows a mercury-cadmium lamp under 
test. The "Man from Mars" helmet 
is, of course, a standard welder's hel- 
met, equipped with special glass to per- 
mit safe viewing of the intense light pro- 
duced by this mercury-cadmium lamp. 
Since this lamp is contained in a quartz 
bulb, it produces high intensities in the 
ultraviolet, so that artificial sunburn is 
difficult to avoid. In studying lamps 
of this type, it is necessary to know as 
much as possible about their color 
quality and variation, if any, in color 
quality as a function of age and various 
operating conditions. Such studies are 
made with a spectroradiometer and 
filtered light meters, and also photo- 
graphically. 

Studies of the zirconium arc, both en- 
closed and open-air varieties, have been 
carried on, although for set lighting pur- 
poses these arcs do not appear to have 
sufficient intensity or satisfactory color 
temperature. 

The xenon gas arc has long been 
known and studied and is perhaps most 
familiar to us in the flashtubes so suc- 
cessfully used for stroboscopic high- 
speed photography. Not so well known 
is the fact that in Germany and England 
development work has been in progress 
on a high-intensity xenon arc of capaci- 
ties ranging up to 1000 w. In Germany 
an air-cooled lamp of this type has re- 
cently emerged from the research lab- 



180 



February 1951 Journal of the SMPTE Vol. 56 




Fig. 3. Production test setup: sealed-beam versus standard studio lamps. 




Fig. 4. Mercury-cadmium lamp under test. 
Kelley and Wolfe: Research Council Activities 



181 



NO FILTER 

RED A 

BLUE C-5 



10 15 20 25 30 

ANGLE OF OBSERVATION IN DEGREES 



Fig. 5. Typical brightness 

fall-off curves goniopho- 

tometer measurements. 



oratories. It is being watched with 
care and samples will be obtained by 
the Research Council as soon as possible. 
This lamp has better color character- 
istics, having almost a continuum 
throughout the entire spectrum and a 
color temperature of the order of 6000 
K (degrees Kelvin), coupled with instant 
starting. If it can be made commer- 
cially available, it can occupy a position 
of real importance in set lighting for 
motion pictures. 

Composite Photography 

Composite photography is a matter 
of vital importance to the motion pic- 
ture industry. It permits making many 
shots which would otherwise be impos- 
sible, and making many more shots 
which would be impractical from an 
economic standpoint if made by any 
other process. There are two general 
types of composite photography, com- 
monly called transparency process pho- 
tography and matte photography. Both 
of these forms of composite pho- 
tography are under study. Some of the 
work done on process screens was re- 
ported to the Society at the Fall Con- 
vention in 1949, 3 but much additional 
work in this field has been done since 
that time. 

A goniophotometer built for our 
special application has been used to 
measure the color characteristics of 



transparency screens. The results of 
one such test are shown in Fig. 5. These 
tests are verified, wherever that is im- 
portant, by actual photographic meas- 
urements, since it must be constantly 
borne in mind that the characteristics 
of the photographic emulsion are an 
inseparable part of the problem. The 
difference in fall-off characteristics of 
this particular screen sample at the dif- 
ferent ends of the spectrum is of obvious 
importance for color photography, but 
is also important for black-and-white 
photography since it must effect the 
resultant definition in many cases. 

The method of making a composite 
photograph which consists of photo- 
graphing foreground objects while simul- 
taneously rephotographing from a screen 
the desired background, can, of course, 

FRONT PROJECTION 




Fig. 6. Schematic drawing 
of front projection setup. 



182 



February 1951 Journal of the SMPTE Vol. 56 




Fig. 7. Front projection with foreground lighting. 




Fig. 8. Front projection without foreground lighting. 

Kelley and Wolfe: Research Council Activities 183 



be employed with a reflection type of 
screen and front projection as well as 
with a translucent screen and rear pro- 
jection. For example, in Fig. 6 is a sim- 
plified setup showing a camera, C, a 
projector, P, and a diaphone mirror, 
M. The picture from the projector is 
reflected by the mirror to the screen, 
S, and rephotographed by the camera 
along with the foreground object. 
Figure 7 is an example of this type of 
photography, for the young lady is 
seated hi front of what appears to be an 
open window through which the city 
may be seen. Figure 8 shows what 
happens if the foreground lights are 
turned off so that the camera sees only 
the silhouette and the rephotographed 
view of the city. This last slide is in- 
cluded primarily to show that the in- 
tensity of light required from the pro- 
jector is insufficient to register on the 
foreground object even though it is 
sufficient to provide a brilliant picture 
of the background. The differences in 
reflection characteristics are, of course, 
responsible for the operation of such an 
arrangement. There are many prob- 
lems in connection with the successful 
use of front projection. The idea is 
not new, but its application and limi- 
tations have never before been properly 
denned, which is the primary object of 
the investigation in that field. 

The industry has long been intrigued 
by the considerable increase in efficiency 
which can be obtained with a directional 
translucent screen as contrasted to a 
nondirectional screen, but in most cases 
the requirement for a mobile camera, 
coupled with manufacturing problems, 
has prevented the use of such screens. 
General awareness of the difficulties and 
the problems involved in a directional 
screen and acquaintance with much of 
the earlier work that has been done on 
this subject have also stimulated the in- 
vestigation hi that direction. There 
presently seems some promise of obtain- 
ing a directional screen which will per- 
mit of camera movement and yet offer a 



light gain of four or five times that pres- 
ently available with the nondirectional 
screens. 

Traveling matte composite photog- 
raphy presents many difficult prob- 
lems. Presently, it is used in the in- 
dustry only where there is no other way 
of making the required picture. This is 
true because the process is slow, ex- 
pensive and it is difficult for many people 
to understand and appreciate the re- 
sults which can be obtained. The Re- 
search Council, in undertaking an in- 
vestigation of this process, expects, 
therefore, to work toward a system 
which will overcome all three of these 
objections. It is hoped to develop a 
system which will be fast and inex- 
pensive and will permit the director, 
cameraman and others concerned to see 
the composite result at the time the 
foreground is being photographed. This, 
of course, can be true only if the back- 
ground material is already available on 
a motion picture film. That's a rather 
ambitious undertaking because it in- 
volves problems of optics, photo- 
graphic materials, lighting and elec- 
tronics. Preliminary studies, however, 
lead to the belief that these highly de- 
sirable results can be achieved. The 
expected improvements in this rather 
old form of composite photography ap- 
pear possible because of improvements 
which have been made in photographic 
film base, emulsions and electronic de- 
velopments. 

The use of projected still backgrounds 
has long been quite a problem, particu- 
larly where color is involved, much of 
the difficulty arising from the instability 
of the colors under the high temperature 
and ultraviolet light conditions which 
prevail. A further difficulty has been 
the problem of matching the foreground 
and background colors, since the fore- 
ground is an original and the back- 
ground is a dupe. These difficulties 
were demonstrated with a frame of a 
35-mm color print [which accompanied 
this paper but cannot be reproduced in 



184 



February 1951 Journal of the SMPTE Vol.56 



color in the JOURNAL] in which the lower 
left-hand quadrant is a direct photo- 
graph of a color chart, and the other 
three quadrants are occupied by pro- 
jected reproductions of the same color 
chart. Two of these are still projec- 
tions and the third is a motion picture 
projection. While none of these match 
the original, it is noted that the diffi- 
culty is principally in the red end of the 
spectrum. This is indeed fortunate since 
still background scenes rarely contain 
any significant red. Colors in such 
scenes are predominantly blue and green, 
where the comparison is not so odious. 
Nevertheless, this is not a satisfactory 
situation and it is hoped that new color 
films which will shortly be on the market 
will correct or at least improve this 
situation. 

Transportation 

Transportation between studios and 
location is a matter of considerable im- 
portance. When the Research Council 
started to analyze this problem, at the 
specific request of the production man- 



agers, it became apparent immediately 
that the problem was incapable of solu- 
tion unless a reasonable amount of 
standardization of equipment required 
for a location could be achieved. Most 
of the equipment required for the aver- 
age location is classed as "grip equip- 
ment" and an analysis covering loca- 
tions made by each of the member 
studios of the Research Council for the 
last ten years developed a definite 
pattern which, after careful study and 
consultation with the grip and camera 
departments as well as a group of 
outstanding directors of cinematog- 
raphy, was consolidated into a standard 
group of equipment. This was classed 
as the basic list to be specified for loca- 
tion unless specific approval for changes 
was obtained from the production 
office. 

A semitrailer was then designed about 
this basic list of equipment with pro- 
visions for carrying all of the grip 
equipment, sound equipment, camera 
equipment and in many cases additional 
equipment as required for the electrical 




Fig. 9. First model of Research Council transportation unit. 
Kelley and Wolfe: Research Council Activities 



185 




Fig. 10. Final design of "standard" 
location transportation unit. 

department, property department or 
others. Figure 9 shows a model which 
was made up to permit graphic discus- 
sion of this problem with all of the de- 
partments involved. Subsequently, the 
design was changed as shown in Fig. 10. 
The folding crane, which is attached to 
the trailer, can be extended and used to 
load or unload any of the equipment 
carried by the semitrailer. It obviously 
can also be used in many other opera- 
tions on location. 

One of the important problems in de- 
signing this unit arose from the variety 
of state laws controlling the size of ve- 
hicles traveling over the roads. Dimen- 
sions were finally chosen to meet the re- 
quirements hi most of the states of the 
Union. California's neighbor to the 
north, Oregon, offered the most diffi- 
cult restriction in a height limitation of 
11 ft. The standard design calls for a 
semitrailer having a height of 12 ft 6 in., 
but the front top of the semitrailer can 
be reduced in height to 11 ft, and if the 
bows are omitted from the balance of 
the top of the semitrailer, the complete 
unit can stay under the 11-ft limitation 
of the State of Oregon, this, of course, 
being accomplished at some sacrifice of 
carrying capacity. In most cases it is 
hoped that a waiver of this restriction 
can be obtained. 

The study and tests which the Re- 
search Council has made on the very 
important problem of film perforations 
was covered in a separate paper pre- 
sented at the same Convention. 4 

Wind Machines 

There are many applications in the 
production of motion pictures for an 
artificial and controlled wind. These 
requirements vary from hurricane con- 
ditions to a gentle zephyr which blows 
milady's scarf. The hurricane in the 
past has been created by airplane 
motors with airplane propellers. In 
many cases, these motors have been 



186 



February 1951 Journal of the SMPTE Vol. 56 




Fig. 11. 6-ft wind machine. 





Fig. 12. Effect of windstream 20 ft from the 6-ft wind machine. 

Kelley and Wolfe: Research Council Activities 187 



Liberty motors of the World War I 
vintage. Since an airplane propeller is 
not designed to create wind, this de- 
vice, while moderately effective, was not 
controllable to a sufficient degree and 
was both noisy and expensive to oper- 
ate. 

Figure 11 shows a machine developed 
by the Research Council for the hurri- 
cane type of application. The fan blade 
is 6 ft in diameter and is a standard unit. 
The motor is a standard industrial gas 
engine, developing approximately 150 
hp. Provision is made for mounting 
two of these motors, one on each side of 
the fan, since about 300 hp is required 
to drive the fan to full output. This 
fan may be tilted through a range of 
from 15 below horizontal to 20 above 
horizontal. The operator has control 
of the speed of the fan and can rotate 
the entire unit through 360. 

Some difficulty has been experienced 
in measuring the velocity of the wind 
created by this machine, but Fig. 12 
shows graphically the intensity that is 
possible when the fan is driven at 200 
hp. The man in the left foreground is 
directly in front of the wind machine at 
a distance of approximately 20 ft and is 
unable to move closer to the fan. The 
sharpness of the beam of wind is shown 
by the fact that the two men in the right 
foreground are almost completely out of 
the air stream. 

Figure 13 shows the wind machine 
being used to blanket a set with smoke. 
The smoke candles can be seen at the 
rear of the fan. 

For use inside the stage during dialog 
sequences, a somewhat similar wind 
machine, having blades 3 ft in diameter, 
was designed and this unit is shown in 
Fig. 14. Like the larger wind machine, 
this unit can be panned or tilted as de- 
sired. It is also designed to permit 
separating the fork supporting the fan 
unit itself from the base so that it can 
be mounted hi a suitable socket on a 
parallel or on scaffolding. It is driven 
by a d-c motor at speeds usually in the 



range from 100 to 400 rpm. Wind ve- 
locities of the order of 8 to 12 mph can 
be created at a distance of 20 ft with a 
noise of about 30 db as measured by a 
General Radio noise level meter on the 
40-db scale. 

A still smaller unit has been de- 
signed, also shown in Fig. 14, again fol- 
lowing the basic impeller principle. 
This unit has a blade which is 18 in. in 
diameter, but is somewhat similar in 
many other respects to the 36-in. fan. 
It is interesting to note that, neglecting 
the noise made by the driving motor, 
the smaller the fan the more noise it 
makes for a given velocity. Thus, if all 
three fans are driven by comparable 
electrical motors, the 6-ft fan is con- 
siderably quieter than the 36-in. fan, 
and it, in turn, is quieter than the 18-in. 
fan. 

Motion Picture Sets 

Most motion picture sets are as- 
sembled from hard flats which consist of 
plywood panels on light wood frames. 
These flats are used over and over 
again. Such sets are almost always 
covered with paper; if a wallpaper dec- 
oration is desired, the wallpaper is 
pasted directly to the flats; if a painted 
surface is desired, a blank paper is first 
pasted on the wall and is then covered 
with paint. Thus, in either case it is 
necessary to remove this paper to pre- 
pare the flats for reuse. In the past this 
has been done by various laborious 
methods, one of which is shown in Fig. 
15. Such procedures were not only ex- 
pensive in terms of time required to re- 
move the paper, but invariably damaged 
the surface of the flat, requiring refinish- 
ing and necessarily reducing the life of 
the flat. 

The Research Council approached 
this problem in two ways. The first 
was the use of the so-called "Peel-Coat" 
which is a paint-like material most com- 
monly known for its use in the storage 
of military equipment, airplanes, ships, 
etc., where a cocoon of this material is 



188 



February 1951 Journal of the SMPTE Vol. 56 




Fig. 13. Making smoke with the 6-ft wind machine. 




Fig. 14. 18- and 36-in. wind machines. 
Kelley and Wolfe: Research Council Activities 



189 




Fig. 15. Old style method of removing wallpaper from flats. 



used to cover the device to be protected. 
This procedure worked with entire satis- 
faction, since pigments could be mixed 
into the Peel-Coat or a coat of paint 
could be applied over the Peel-Coat, or 
paper could be applied over the Peel- 
Coat, and hi each case the surface could 
be stripped easily by the simple oper- 
ation of making an incision through the 
Peel-Coat with a knife or other sharp 
instrument and then stripping the whole 
business off the flat. However, the 
process was expensive and did not meet 
with favor, and as a result, efforts were 
concentrated on a simpler solution. 
This came out in a material known as 
Peel Paste which was developed en- 
tirely by the engineers of the Research 
Council. It has the consistency of 
ordinary wallpaper paste and permits 
the paper to be worked in exactly the 
same manner as the more familiar wall- 
paper paste. It holds the paper on the 
wall satisfactorily until it is time to re- 
move it, when again an incision is made 



with a knife and the entire section of 
paper is stripped, usually in one piece. 
Figure 16 shows such an operation being 
performed on a standing set. 

Similar strippable adhesives have 
been developed for use with various 
types of floor covering; rubber tile, 
asphalt tile, linoleum, parquet floors 
and similar material. Here, the strip- 
pable adhesive permits removing the 
floor covering without damage to either 
the floor covering itself or to the floor 
on which it is laid. 

Like most other industries, motion 
picture studios have been making in- 
creased usage of plastics. If one wishes 
to be technical, it might be pointed out 
that the motion picture industry de- 
pends entirely on plastics since the film 
base is in itself a plastic material, but 
there are also a host of other important 
usages of plastics, many of which have 
grown up since the close of World War 
II. The Research Council is constantly 
testing, in cooperation with the studios, 



190 



February 1951 Journal of the SMPTE Vol. 56 




Fig. 16. New procedure for 
stripping wallpaper from flats. 

new applications of plastics. These in- 
clude such things as flexible molds for 
casting plaster, hardening materials 
to be mixed with the plaster, thermo- 
plastic materials for set construction 
items such as stair handrails, plastic 
props, simulating metal with plastic 
as in the armor worn by knights, a full- 
scale locomotive and many others. 

It is presently standard practice to 
move set walls from the mill to the stage 
and from the stage to the scene docks on 
wheeled platforms called set dollies. 
These little platforms are made up of a 
couple of 2 X 12's, 3 or 4 ft long, with 
four 2- or 3-in. free casters. They are 
inexpensive and their very simplicity 
makes them extremely versatile, but 
they also have some distinct shortcom- 
ings. Their stability is not good and the 



small casters frequently catch in holes 
in the ground. As a result, the Re- 
search Council has undertaken the de- 
sign of an improved dolly. 

All of the major studios own rather 
extensive, permanent outdoor sets, 
mostly in the form of streets of one kind 
or another. If any extensive shooting 
is to be done on these sets, it is usually 
necessary to cover them in with canvas 
so that the direction, intensity and color 
quality of the light will always be under 
the control of the cameraman. This 
means that each of the studios has liter- 
ally acres of canvas which they call 
diffusing cloths. These diffusing cloths 
must be flameproofed to minimize the 
fire hazard, and this flameproofing treat- 
ment increases their weight and makes 
them more difficult to handle. With 
or without the flameproofing treatment, 
the life of canvas exposed to the at- 
mosphere of Los Angeles is relatively 
short, perhaps two or three years. 

The Research Council is, therefore, 
in the business of studying fabrics. 
First it seemed that nylon would be a 
natural answer since it was already 
flameproof and known to be consider- 
ably stronger than cotton fabrics such 
as canvas. Tests very quickly proved 
that this was not the answer, as nylon 
will not stand up under these atmos- 
pheric conditions as well as canvas 
does. Glass cloth is, of course, fireproof 
and stands up well under atmospheric 
conditions, but is easily damaged by 
abrasion and in many cases its tear re- 
sistance is low. There is no answer to 
this problem at the moment, but tests 
are in progress on nylon and glass 
cloth, each with a vinyl plastic coating. 

Camera Cranes and Dollies 

Some time ago the Research Council 
developed a camera crane which was 
reported to this Society in a paper pre- 
sented at the 1948 Fall Convention. 5 
Since that time there has been designed 
a dolly for this camera crane, as shown 
in Fig. 17. Pneumatic tires are em- 



Kelley and Wolfe: Research Council Activities 



191 




Fig. 17. Camera crane dolly. 



ployed and to avoid the necessity for a 
differential, two series d-c motors are 
used to drive separately two of the 
wheels of the dolly. Steering, acceler- 
ation and brake have been patterned 
after those hi an automobile hi order 
that the operator may feel at home. 

Figure 18 shows a camera crane 
mounted on the camera-crane dolly. 
The operation of mounting or removing 
the camera crane from this dolly can be 
accomplished without any special tools 
and without a hoist. It is but a matter 
of a few minutes' work. 

Figure 19 illustrates a camera geared 
head whose design is quite different 
from those commonly used hi the in- 
dustry. It permits tilting the camera 
through an arc of 45 each side of the 
horizontal and thus has some distinct 
advantages over other geared heads. 

Pho tography 

Figure 20 shows a doorway at night 
with a light shining through the glass 
door panels. Actually, the door panels 
are made of a highly directional reflect- 



ing material and the light used was a 
small spotlight located on the camera. 
This is an example of a simple applica- 
tion of readily available materials which 
can be used to good advantage in this 
industry. 

Although a picture is photographed on 
a two-dimensional medium (the film 
itself) and projected on another two- 
dimensional medium (the theater 
screen), the industry has always wanted 
a picture in three dimensions. There 
have been a number of papers before 
this Society with demonstrations of 
systems which permit of all three di- 
mensions. Some of these have em- 
ployed polarized light and others have 
obtained their separation by color, and 
similar procedures, but in every case 
they require the use of some kind of 
crutch by each individual hi the audi~ 
ence, or they restrict the viewer's posi- 
tion and motion of his head in a most 
unnatural way. So far the industry 
has been unwilling to make any com- 
mercial use of any of these systems, ex- 
cept on a novelty basis. 



192 



February 1951 Journal of the SMPTE Vol. 56 




I 



Fig. 18. Camera crane mounted on the dolly. 




Fig. 19. Geared head. 
Kelley and Wolfe: Research Council Activities 



193 




Fig. 20. "Scotchlite" window reflector. 



The Research Council is constantly 
receiving proposals from inventors all 
over the world for systems to permit 
three-dimensional motion pictures. So 
far none of these systems appears prac- 
tical. Nevertheless, each one is care- 
fully considered and investigated if 
that seems necessary. In order to 
understand better the problems of 
three-dimensional motion pictures, the 
Research Council has purchased an 
attachment for a 16-mm camera, as 
shown in Fig. 21, which permits photo- 
graphing a stereo pair on the film. Fig- 
ure 22 shows what this stereo pair looks 
like on the film. It is turned on its side 



to permit maximum usage of the film 
area and when it is projected through 
the same attachment used in making the 
picture, plus a polarizing screen, and 
viewed with proper analyzing glasses, 
a motion picture in three dimensions is 
obtained which is satisfactory for lab- 
oratory invest igational purposes. 

The Research Council activity in 
connection with color is largely con- 
fined to reporting to our member com- 
panies on various color systems as they 
are announced and studying problems 
of test and control for color systems 
which seem likely to receive commercial 
usage. We are consequently interested 



194 



February 1951 Journal of the SMPTE Vol. 56 




Fig. 21. 16-Mm stereo camera. 




Fig. 22. Still shot through stereo attachment. 
Kelley and Wolfe: Research Council Activities 



195 



in color densitometers, color charts, 
printing machines and similar devices. 

Magnetic Recording 

In the field of magnetic recording 
and in the older art of photographic re- 
cording, the Research Council has not 
been particularly active because both 
the studios and their suppliers are ac- 
tively at work on these problems. An 
analysis of the economic problems which 
needed consideration in connection 
with magnetic recording was prepared, 
however, because the differences in op- 
erating practices and requirements 
throughout the industry were creating 
false impressions which needed correc- 
tion. 

Television 

Television presents another situation 
where the Research Council can only 
hope to keep abreast of that fast- 
changing art so that its member com- 
panies may be advised when television 
systems, equipment or techniques reach 
the place where they can be profitably 
applied to the production of motion 
pictures. In other words, the Research 
Council is not concerned with television 
as a medium of home entertainment. 
It is concerned with it as a medium of 
theater entertainment and as a means 
of producing motion pictures. 

There are many other relatively minor 
items in which the Research Council is 
active. They include, for example, prob- 
lems of flicker, elimination of static on 
film, special types of storage batteries, 
new microphone booms and refriger- 
ated film-shipping containers. In fact, 
the Research Council is interested in 



anything which has an application as a 
tool in the making or exhibition of a 
motion picture. 

There is oftentimes some confusion 
regarding the relationship of the Motion 
Picture Research Council to the Society 
of Motion Picture and Television Engi- 
neers. This misunderstanding usually 
arises from matters having to do with 
either standards activities or test films. 
The Research Council works very 
closely with the Society on all problems 
of standardization within the motion 
picture industry, but as a member body 
of the American Standards Association, 
the Research Council also acts directly 
on such problems. The Society and the 
Research Council work very closely to- 
gether in the test-film field, each accept- 
ing orders for test films made by the 
other. Test films are looked upon as a 
service to the exhibition end of the in- 
dustry which has been undertaken to 
insure satisfactory presentation of the 
studio product in the theater. 

References 

1. W. F. Kelley, "Motion Picture Re- 
search Council," Jour. SMPE, vol. 
51, pp. 418-423, Oct. 1948. 

2. Wayne Blackburn, "Study of sealed 
beam lamps for motion picture set 
lighting," Jour. SMPTE, vol. 55, pp. 
101-1 12, July 1950. 

3. Herbert Meyer, "Sensitometric aspects 
of background process photography," 
Jour. SMPTE, vol. 54, pp. 275-289, 
Mar. 1950. 

4. W. F. Kelley and W. V. Wolfe, "Re- 
cent studies on standardizing the 
Dubray-Howell perforation for uni- 
versal application," Jour. SMPTE, 
vol. 56, pp. 30-38, Jan. 1951. 

5. Andre Crot, "Research Council small 
camera crane," Jour. SMPE, vol. 52, 
pp. 273-279, Mar. 1949. 



196 



February 1951 Journal of the SMPTE Vol. 56 



Semiautomatic Color Analyzer 

By Lloyd E. Varden 



A semiautomatic color analyzer is described for rapidly determining the 
extent of unbalance, or deviation from "type," of a processed color negative 
or color positive monopack film. A standardized light source, a rotating 
color filter having three sectors which transmit narrow spectral bands of 
blue, green and red light, a multiplier-type phototube and amplifier, and 
a cathode-ray tube are employed. The sweep circuit of the cathode-ray 
tube is synchronized with the rotating filter wheel so that a horizontal 
straight-line image is produced when a gray or near-gray density of a 
"balanced" sample is in the light path. A cathode-ray tube image which 
deviates from a horizontal straight-line image indicates unbalance in a 
test sample, whereupon correction filters can be introduced in the light 
path by means of servomechanism devices to produce a horizontal straight- 
line or "balanced" condition. 



IT is WELL KNOWN that for many prac- 
tical purposes a complete energy 
versus wavelength relationship is not 
always required to determine the ef- 
fective spectral characteristics of a 
light source or of a selective absorbing 
material in terms of some adopted 
standard. Tungsten lamps, for ex- 
ample, can be calibrated against a 
black-body radiator or a suitable 
secondary standard, and any deviations 
in spectral emission from lamp to lamp 
can be expressed as color temperature 
differences or as voltage differences 
necessary to produce a constant color 
temperature. The ratio of only two 
values of a lamp's spectral emission, one 
in the blue region and one in the red 
region, is sufficient for such a specifica- 
tion. 

Presented on October 17, 1950, at the 
Society's Convention at Lake Placid, 
N.Y., by Lloyd E. Varden, Pavelle Color 
Inc., 533 W. 57th St., New York 19, 
N.Y. 



Similarly, simplified methods are 
utilized to express the spectral absorp- 
tion characteristics of processed multi- 
layer color films. Integral density 
measurements made at only three 
spectral positions at wavelengths cor- 
responding with the spectral absorption 
peaks of the dyes formed in the layers 
are used for most laboratory control 
purposes. For color sensitometry, the 
more meaningful equivalent gray den- 
sities are preferred. These can be 
measured directly or can be derived from 
the integral densities. Such density 
measurements, as well as the entire 
subjects of color densitometry and color 
sensitometry, are summarized in the 
recent report of the Color Sensitometry 
Subcommittee. 1 The only point to be 
stressed here is that for each density 
specification three measurements are 
required. This can become a bottle- 
neck in laboratory practice if numerous 
samples must be read. To use these 
values for setting up color correction 



February 1951 Journal of the SMPTE Vol. 56 



197 



filter combinations is further time- 
consuming since the densities obtained 
must be correlated with the proper cor- 
rection filter densities. 

In the instrument to be described, a 
method is provided for determining 
very rapidly the correction filters neces- 
sary for printing a color film, assuming 
that all scenes will appear satisfactory 
when corrected to the same standard. 
It is recognized that any given stand- 
ard, for example, gray, may not result 
in the most pleasing results for all 
scenes. Nevertheless, a gray balance 
condition as a first approximation is 
desirable, especially in color negative- 
color positive processes, since it will 
generally give the most acceptable over- 
all quality, and is the best starting point 
for making changes if any are indicated. 

The instrument is of most value for 
color negative-color positive processes, 
because in these processes it is difficult 
to estimate visually what correction 
filters are necessary from the color 
negative images. Color positive images 
can be evaluated fairly well by visual 
methods, but even here large errors are 
possible, especially if the color of the 
image deviates appreciably from normal. 

Description of the Instrument 

The principal errors in the color bal- 
ance of a reproduction arise from: 

1. Light source color-quality vari- 
ations, 

2. Film color-balance variations, and 

3. Processing variations. 

The combined effects of these can be 
measured from control densities placed 
on the film for this purpose. One or 
more gray values can be photographed 
at the beginning of each scene to estab- 
lish the control density at or near the 
middle of the density scale of the film. 
For convenience we can assume that 
the reproduction of these gray values 
should be gray if the light source, film 
color balance and processing are normal. 
(This may not be true, however, for 
intermediate reproduction steps even 



if gray reproduction has been accepted 
as normal for the final image.) 

The problem, then, becomes one of 
determining whether or not the repro- 
duction of gray is correct, and if not, 
expressing the deviation from gray in 
terms of correction filters required in 
printing to restore the gray condition. 

Figure 1 is a schematic view of an 
instrument designed for this purpose. 
Its principle of operation is as follows: 
A standardized light source is focused 
upon the cathode of a multiplier photo- 
tube, the amplified output of which is 
connected across the vertical plates of a 
cathode-ray tube. In the light path 
immediately above the phototube is 
a rotating filter wheel containing three 
sectors. Each sector transmits but 
one narrow spectral band in the blue, 
green or red region. The filter com- 
binations used to isolate these bands are 
the same as those in the original 
models of the Ansco Color Densi- 
tometer, giving transmission peaks at 
440 m M , 540 m M and 660 m M for the dif- 
ferent filter sectors. 2 The rotation of 
the filter wheel is controllable from 300 
to 900 rpm, but at any speed is syn- 
chronized with the horizontal sweep of 
the cathode-ray tube so that the first 
third of the tube pattern corresponds 
to the blue filter, the middle third to 
the green filter and the remaining third 
to the red filter.* Synchronization is 
accomplished by means of a small 
Alnico magnet on the periphery of the 



* Numerous instruments have been de- 
scribed for various spectral analysis pur- 
poses which employ rotating filters or 
other wavelength isolation means in 
conjunction with a photocell and cathode- 
ray tube. An equivalent gray color den- 
sitometer having such components was de- 
scribed by Senger, 1 Schneider 4 and 
Schneider and Berger. 6 Typical of elec- 
tronic spectrographic equipment are the in- 
struments of Feldt and Berkley, 6 - 7 Dieke 
and Crosswhite 8 and Sziklai and Schroe- 
der. 9 - 10 Zworykin and Ramberg 11 give 
a general discussion of the subject. 



198 



February 1951 Journal of the SMPTE Vol. 56 



wheel. This forms an electrical im- 
pulse in a coil which is amplified and 
used to trigger the sweep of the cathode- 
ray tube. 

The instrument is first balanced with 
a "type" sample in the light path by 
placing neutral densities in the filter 
sectors until a horizontal straight-line 
pattern is obtained on the cathode-ray 
tube. With this condition established, 
an off-gray sample substituted for the 
"type" will cause amplitude changes in 
the cathode-ray pattern depending upon 
the relative change it brings about in 
the amount of blue, green and red light 
reaching the phototube. For example, 
a sample deficient in magenta will allow 
an excess of green light to pass relative 
to the blue and red light transmitted. 
Therefore, the pattern on the cathode- 
ray tube will no longer be a horizontal 
straight line, but will rise principally 
in the middle section corresponding to 
the "green" position of the rotating 



filter wheel. However, no part of the 
line image remains unaffected because 
the secondary absorptions of the ma- 
genta dye in the blue and red regions 
are also lacking. 

Above the rotating filter wheel is a 
stack of three filter correction wheels, 
each having five openings. One open- 
ing in each wheel has no filter. This 
position, of course, is used when the 
instrument is balanced, or when an 
unknown sample is first placed in the 
light path. Different densities of yel- 
low, magenta and cyan filters are in the 
other openings of the wheels, one color 
series in each wheel.* The dyes used 

* Four filter densities and one blank 
in each of the correction filter wheels 
were found to be insufficient to meet all con- 
ditions in practice. Therefore, the instru- 
ment now has been revised to allow for 
several thousand filter combinations by 
adding three additional filter wheels, 
each having 14 apertures. 



TO POWER 
"UPPLY 



SCHEMATIC DIAGRAM 

OF 

SEMI -AUTOMATIC 
COLOR ANALYZER 



COMPENSATING FILTER 
WHEELS 

SERVO-MOTORS , ROTAT 




Fig. 1. Schematic view of semiautomatic color analyzer. 
Lloyd E. Varden: Color Analyzer 



199 



for these filters are the same as those 
which form in the layers of the color 
film. 

The positioning of the various densi- 
ties of the correction filters is accom- 
plished by a servomechanism system, 
consisting of a series of pushbutton 
switches for each wheel, a synchro 
system, a servo amplifier and servo- 
motor. When a given button is pushed 
down, the motor rapidly turns the 
wheel to place the correct filter in the 
light path as determined by the preset 
synchro system. The last button 
pushed down stays down so that the 
filters required for printing can be noted 
after balance has been established. 

It is clear that yellow, magenta or 
cyan deficiencies of a sample can be 
ascertained by introducing the cor- 
rection filters needed to restore a hori- 
zontal straight line on the cathode-ray 
tube. In the previous example, the 
magenta deficiency of the sample is 
compensated by use of a magenta filter 
of the required density. The hump in 
the middle of the tube pattern is flat- 
tened as increasing magenta density 
is introduced, and at the same time the 
apparent deficiencies in yellow and 
cyan disappear. For a color positive or 
color negative material the magenta 
density added in this case indicates 
directly the color and density of filter 
required for printing. 

Figures 2a, 2b and 2c illustrate the 
operation of the instrument. In Fig. 
2a is shown the straight-line, balanced 
condition in which the "type" sample 
is in the light path and all three of the 
correction filter wheels are in the no- 
filter position. When the magenta- 
deficient sample is placed in the light 
path in place of the "type" sample, the 
cathode-ray tube pattern changes as 
shown in Fig. 2b. By introducing the 
proper density magenta filter, the 
straight-line condition is restored, as 
shown in Fig. 2c. The fourth button 
in the middle row of buttons was pushed 
down to re-establish the balance of the 




Fig. 2a. Photograph of cathode-ray 
tube pattern showing straight-line, 
balanced condition when a "type" 
sample is in the light path and all 
three correction filter wheels are set 
in the no-filter position. 

instrument for this particular sample. 

The control density of the sample to 
be analyzed is not critical, but it should 
fall along the straight-line portion of 
the D-logwE curves where it can be 
assumed that the curves for the blue, 
green and red densities are parallel. 
If the curves are not parallel the instru- 
ment will merely indicate the filters 
necessary to pull the curves together 
at one cross-over point. Parallelism 
of the curves can be ascertained with 
the instrument by use of a graded den- 
sity sample. When such a sample is 
balanced for one density and then moved 
slowly from this density level to an- 
other, the cathode-ray pattern remains 






200 



February 1951 Journal of the SMPTE Vol. 56 




Fig. 2b. Photograph of cathode-ray 
tube pattern for a magenta-deficient 
sample. The central hump indicates 
the excessive green light transmitted 
by the sample. 



Fig. 2c. Photograph of cathode-ray 
tube pattern for same sample as used 
for Fig. 2b, except a magenta correc- 
tion filter has been introduced into 
the light path by pressing button in- 
dicated by arrow. 



a straight-line if the curves are parallel. 
If the curves are not parallel, each 
change of density in the sample will 
require different balancing filters. 

Figure 3 shows an over-all view of 
the instrument. The pushbuttons for 
positioning the correction filters, the 
cathode-ray tube screen and the aper- 
ture for placing the sample are in line, 
one above the other, with the cathode- 
ray tube placed at an angle for con- 
venience in observing the pattern. At 
the sample position a sliding tube is 
provided to uncover the aperture for 
inserting the sample in the light path. 
When this tube is in the "up" position 
the aperture is illuminated from be- 



neath so that the film density to be 
evaluated can be situated properly. 
The tube is then lowered over the sample 
to cut out extraneous light and to secure 
the film. 

Figure 4 shows a close-up view of the 
control panel. The dials in the back are 
for presetting the synchro system so 
that the openings of the filter correction 
wheels fall in the light path. A stand- 
by switch is provided so that the 
power supply can remain on when the 
instrument is not in use. A speed ad- 
justment for the rotating filter wheel 
is essential to obtain a smooth, steady 
trace on the cathode-ray screen. The 
usual cathode-ray tube controls are 



Lloyd E. Varden: Color Analyzer 



201 




Fig. 3. Over-all view of color analyzer. 




202 



Fig. 4. Close-up view of control panel. 
February 1951 Journal of the SMPTE Vol. 56 




Fig. 5. Internal view of rotating filter mechanism, correction filter wheels, 
photomultiplier tube and servo amplifier units. 




Fig. 6. View of one side of the color analyzer cabinet showing the readily re- 
movable electronic units. 

Lloyd E. Vardeii: Color Analyzer 203 



also provided for positioning the trace 
in the % center of the screen, for increas- 
ing or decreasing the brightness of the 
image, for focusing and for controlling 
the amplitude response. It is seldom 
necessary to use any of these controls 
once the instrument has been adjusted. 

Figure 5 shows the internal mecha- 
nism of the filter correction wheels, the 
rotating filter wheel, the servomotors, 
etc. Also shown in Fig. 5 is an inside 
view of two of the servo amplifier units. 
These are standard units manufactured 
by Servomechanisms, Inc., Mineola, 
L.I., and are readily removable for 
inspection. 

One side of the instrument is shown 
in Fig. 6 to illustrate how the electronic 
components are fitted into the cabinet 
for convenient servicing. 

Acknowledgments: The following 
people have given valuable assistance 
in the final design, construction or 
testing of the instrument: Dr. Herman 
Duerr, Monroe H. Sweet and John 
Forrest of Ansco; William Shannon 
and Ralph Redemske of Servomecha- 
nisms, Inc.; Leo Pavelle, Peter Krause 
and Rudy Seefried of Pavelle Color 
Inc. 



References 

1. "Principles of color sensitometry 
a report of the Color Sensitometry 
Subcommittee," Jour. SMPTE, vol. 
54, pp. 653-724, June 1950. 

2. M. H. Sweet, "An improved photo- 
multiplier tube color densitometer," 
Jour. SMPTE, vol. 54, pp. 35-62, 
Jan. 1950. 

3. N. Senger, "Progress in the field of 
the Agfacolor process," Film und 
Farbe (Dresden, 1942), pp. 11-14, 
Max Hesses, Berlin, 1943. 

4. W. Schneider, "The Agfacolor proc- 
ess," FIAT Final Report No. 976, 
n.d. 

5. W. Schneider and H. Berger, "On the 
sensitometry of the Agfacolor proc- 
ess," Zeits. /. wiss. Photographic, 
Photophysik und Photochemie, vol. 42, 
pp. 43-52, 1943. 

6. R. Feldt and C. Berkley, "The 
cathode-ray spectrograph," Proc. Nat. 
Electronics Con/., pp. 198-211, Oct. 
1946. 

7. R. Feldt and C. Berkley, U.S. Pat. 
2,444,560. 

8. G. H. Dieke and H. M. Crosswhite, 
"Spectrochemical analysis with the 
oscillograph," /. Opt. Soc. Amer., 
vol. 36, pp. 192-195, 1946. 

9. G. C. Sziklai and A. C. Schroeder, 
"Electronic spectroscopy," J. Appl. 
Phys., vol. 17, pp. 763-767, 1946. 

10. G. C. Szirlai and A. C. Schroeder, 
U.S. Pat. 2,519,154. 

11. V. K. Zworykin and E. G. Ramberg, 
Photoelectricity and Its Application, 
Wiley & Sons, New York, 1949. 



204 



February 1951 Journal of the SMPTE Vol. 56 



Motion Picture Studio Lighting Committee Report 



By M. A. I la 11 kitis. Committee Chairman 



During the past several years the Motion Picture Studio Lighting Com- 
mittee reports and papers have described studio lighting equipment, set 
power distribution and power supply. 1 " 6 Some mention has been made 
of set lighting levels and lamp location but the variables are so great it 
was found exceedingly difficult to provide the information within the 
scope of a paper or report. 

Because of numerous requests for at least a general picture of set light- 
ing levels and equipment placement, this report will describe and illus- 
trate representative sets which were lighted for three-color photography 
and come within what may be termed as "high-key" lighting. 



IN SET LIGHTING for motion pictures the 
cinematographer thinks more in 
terms of obtaining an emotional effect 
that will carry the mood of the picture 
to the audience than upon correct 
exposure alone. If he must make a 
choice between working within the 
normal latitude of the film or obtaining 
the best dramatic effect, he will usually 
choose the latter course. Whether 
or not his result is satisfactory is a 
measure of his combination of artistic 
and engineering ability. 

From an engineering viewpoint he 
may find it desirable to establish his 
key-light in the middle range of the 
latitude of the film and to restrict high- 
light and shadow areas to a ratio that 
will assure correct exposure. From 
an artistic viewpoint, however, he 
may not be able to obtain the dramatic 
effect for which he is striving and he will 
experiment outside of engineering limits 



Presented on October 19, 1950, at the 
Society's Convention at Lake Placid, 
N.Y. 



for the best combination of exposure 
and dramatic effect. 

The key-light levels on the color 
sets described vary from 500 to 600 
ft-c. While there may be instances of 
sets which are photographed at higher 
levels than those indicated, for the 
most part the trend would be downward, 
even toward key-light levels as low as 
50 ft-c on some gangster, or mystery 
type, black-and-white pictures. 

A study of the following data will 
cause some to wonder why a set is 
rigged with more lamps than the total 
operating load indicates as having been 
used. The question is answered by the 
fact that when the cinematographer 
starts shooting he must have lamps in 
place for long shots, medium shots, 
dolly shots, and close-ups to avoid the 
necessity of the loss of expensive shoot- 
ing time in moving lamps. 

Quite often the cinematographer sees 
a given set for the first time shortly 
before he starts to photograph the 
picture. He has had little to do with 
the shooting arrangement, color balance 



February 1951 Journal of the SMPTE Vol. 56 



205 




If 

23 



206 



February 1951 Journal of the SMPTE VoU 56 



O.H HANGING 



\ 



\ 



\ 





6 FT. ROLLER 
PLATFORM 



HANGING LAMP 
PLATFORM 



HANGING LAMP 
PLATFORM 



SYMBOL 


ARC LAMPS 





TYPE 170 





" 90 i 


o 


40 j 




INCAN. LAMPS 


s 


SENIOR 


j 


JUNIOR 


B 


BABY 



CAMERA 

Fig. 2. Gaffer's layout of butler's pantry scene in Lullaby of Broadway. 
Courtesy of Electrical Dept., Warner Bros.; Pictures, Inc. 



or general preplanning, yet he must be 
in a position to establish and maintain 
key-light levels on characters who are 
moving about, and often with the 
camera in movement on a dolly as well. 
Furthermore, even after shooting has 
started, he is often called upon to re- 
arrange his lighting for a different 
camera angle than was originally 
planned. 

It would seem, if the latitude of the 
particular color process is to be sacri- 
ficed for dramatic effect, there would 
be little hope of expecting the optimum 



in color quality. In actual practice 
the reverse is true because color quality 
has been steadily improving Jn the face 
of fewer restrictions placed "upon the 
cinema tographer and of lower light 
levels being used. It is merely 'a- ease 
where the end result is dramatic effect 
and engineering ability is being applied 
to make the process meet the needs of 
the end result rather than the apparent 
exposure requirements of the film 
alone. 

It is a virtual impossibility to establish 
hard and fast rules and regulations for 



Studio Lighting Committee Report 



207 




February 1951 Journal of the SMPTE Vol.56 



the lighting of a given motion picture 
set since each cinematographer will 
light a set to satisfy his individual 
artistic interpretation of the dramatic 
effect he is striving to produce. How- 
ever, in order to indicate general set- 
lighting requirements, a survey was 
made of three motion picture sets in 
production: a small set, one of medium 
size and a large one. Information 
concerning how these representative 
sets were lighted for three-color photog- 
raphy is contained in Table I. 

A study of Table I shows that while 
the area of the set in Lullaby of Broad- 
way, Figs. 1 and 2, was only half of 
that in Home on the Riviera, Figs. 3a, 
3b and 4, the total peak load was 
almost the same. This may be ac- 
counted for by the actual area being 
illuminated on the set, by the type of 
lamps needed for the particular effect 
and by the mood of the effect itself. 
In Figs. 1 and 2 a higher key-light level 
is maintained than on the scene shown 



in Figs. 3a, 3b and 4. However, less 
light is needed on the walls to bring 
them into proper perspective with the 
balance of the scene. Also, in Figs. 
1 and 2 the shadow areas are illumi- 
nated, whereas in Figs. 3a, 3b and 4 
they are allowed to go black. 

The Samson and Delilah set, illus- 
trated in Figs. 5 and 6, shows the light- 
ing equipment used on large areas where 
daylight intensity is indicated. It is 
interesting to note that the high light- 
to-shadow ratio was maintained within 
narrower limits than on the other sets 
illustrated. 

References 

1. R. G. Linderman, C. W. Handley and 
A. Rodgers, "Illumination in motion 
picture production," Jour. SMPE, 
vol. 40, pp. 333-367, June 1943. 

2. "Report of the Studio Lighting Com- 
mittee," Jour. SMPE, vol. 34, pp. 
94-97, Jan. 1940. 

(concluded on p. 211} 



Table I. Data on the Lighting for Three-Color Photography of Representatire 
Motion Picture Sets of Small, Medium and Large Size. 

Lullaby of Broadway On the Riviera Samson and Delilah 



Scene Butler's Pantry 


Home Interior 


Temple of Dagon 


Width of set, ft 16 


25 


75 


Length of set, ft 28 


40 


265 


Area of set, sq ft 448 


1000 


19,875 


Height of lamp parallels, ft 14 


13 


25, 30 & 34 


Color of walls light blue - green 


tan 


gravish vellow 


Key-light level, ft-c 600 


500 


550 


Average light level on walls, ft-c 200 
Min. light level in shadow area, 


300 


500 


ft-c 50 


aoprox. zero 


100 


Max. highlight level, ft-c 600 


500 


550 


Camera lens diaphragm opening //2 . 2 


//2-2 


7/1. 9 


Type and num- Type 450 
ber of arc lamps Type 170 6 


7 


30 
232 


available on Tvpe 90 11 


12 


42 


set Type 40 15 


14 


40 


Type and num- Senior 16 


40 


14 


ber of incan- Junior 13 


21 




descent lamps Baby 15 


12 




available on set Sky Pan 




24 


Total paper load, amp 3985 


5550 


57,090* 


Peak load used, amp 2450 


2860 


48,000 


Photographs of set Fig. 1 
Gaffer's layout sketch Fig. 2 


Figs. 3a and 3b 
Fig. 4 


Fig. 5 
Fig. 6 



Studio Lighting Committee Report 



209 




s 

II 



- x 

M 

* 



if 
II 

^ r 

it 



1-3 

^ >* 

"Si 

r 



a 



210 



February 1951 Journal of the SMPTE Vol. 56 



LiLLl'S BEDROOM 



FIRE PLACE FR. DOORS 

! 1 \ / 




o 



SYMBOL 


ARC LAMPS 





TYPE 170 





90 


O 


" 40 




INCAN. LAMPS 


s 


SENIOR 


J 


JUNIOR 


B 


BABY 




5FT. PARALLEL 



Fig. 4. Gaffer's layout of home interior scene in On the Riviera. 
Courtesy of Electrical Dept., Twentieth Century-Fox Film Corp. 



-, Jour. SMPE, vol. 35, pp. Committee Members 



3. - 

607-609, Dec. 1940. 

4. - , Jour. SMPE, vol. 47, pp. 
113-1 18, Aug. 1946. 

5. - , Jour. SMPE, vol. 49, pp. 279- 
288, Sept. 1947. 

6. - , Jour. SMPE, vol. 51, pp 



Richard Blount 
J. W. Boyle 
Karl Freund 
C. W. Handley 



656-666, Dec. 1948. 

Studio Lighting Committee Report 



Petro Vlahos 

C. R. Long 
W. W. Lozier 

D. W. Prideux 



211 







5 fi 

If 
I 



212 



February 1951 Journal of the SMPTE Vol.56 






























(D <D <D <D 
9 







D 










O 
(D <D 






<n o 

a. ON- 

2 *- 

J UJ 

2 > 


i? 


INCAN. LAMP* 
SENIOR 
JUNIOR 

B A BV 


Z 

3> 

I* 

5</> 







<D (E 










j 























ecD 


)O 


tn-a 


10. 







<D C 

8 




a 


> 
























0)(D 




a 







) 






























0.0. 










(D 

: 






c 





) 












" 



w 

o 









OOOOJO.OO O 







O O O 


l 


o o o o 




0.0. 







x. 
o 

UJ 

Q 













00 














<r 




























<r 

UJ 
Q. 
Q. 


OWER 









CD 





(D 









0.0. 

1 





-, 


- 1 


o>OOOOJOOO O 






o o o < 


J | 


o o o o 




o <s 













(D 






Q 




O. O. 










e * 












a 


<D 
i 




1 

(D 
O 










































a 


> 




0.0. 























































0.0. 












d 









3> 




e 














o. a. 


a 


a. a. 




o a. 











n 


) 


































0.0. 












Q g> 


< 


D 









D 

































Fig. 6. Gaffer's layout of Temple of Dagon scene in Samson and Delilah. 
Courtesy of Electrical Dept., Paramount Pictures, Inc. 



Studio Lighting Committee Report 



213 



A High-Speed Stereoscopic 
Schlieren System 



By John H. Hett 



A stereoscope Schlieren system is described. A 16-mm Fas tax Camera is 
used to photograph 10 -in. sections of a 4 by 4 in. flame tube at 9000 
frames/ sec. A polarizing projection system is used. The depth effect is 
observed with burning gases provided sufficient detail is available in 
the image. 



rriHE NORMAL Schlieren system gives 
J_ a flat two-dimensional field, so 
that an observed event cannot be ac- 
curately located along the line of sight. 
To study the details of such phenomena 
as burning jet formation in combustible 
gases, the acceleration of flame fronts 
and the growth and decay of turbulence 
in flames, a stereoscopic system is 
indicated. 

The system shown in Fig. 1 was de- 
signed to observe a working section of 
approximately 5 by 10 in. The two 
large mirrors are parabolae 16 in. in 
diameter by 60 in. in focal length. The 
flat mirrors, A, are 6 in. by 10 in., front 
aluminized. These mirrors should be 



Presented on October 18, 1950, at the 
Society's Convention at Lake Placid, 
N.Y., by John H. Hett, Research Div., 
New York University College of Engi- 
neering, University Heights, New York 
53, N.Y. The work described in this 
paper was done with the support of the 
Office of Naval Research, Dept. of the 
Navy, and the Office of Air Research, 
Dept. of the Air Force, under Contract 
N6ori-ll, Task Order 2, as part of Proj- 
ect Squid. 



flat to better than 2 bands over an 8-in. 
diameter. If this accuracy cannot be 
held, it will be necessary to balance 
the aberrations in the two branches of 
the system, and possibly give special 
shapes to the knife edges. The in- 
cluded angle between the two main 
branches was chosen as 15 at the work- 
ing section, which is approximately the 
angle of convergence of normal eyes at 
reading distance. Higher angles of 
convergence up to 35 were tried with- 
out improving the performance. 

This system is usually used to ob- 
serve 10-in. long sections of a 4 by 4 in. 
square flame tube Figure 2 is a photo 
of the system. Two images are formed, 
one above the other on each frame of the 
Fastax 16-mm camera. The images 
on the film are set so that they are 
about 1 mm apart at the edges. Fig- 
ures 3a and 3b show how the images 
appear on the film. Figure 4 shows the 
test hexagon. 

The light source used is a 100-w zir- 
conium lamp with a horizontal cylin- 
drical condenser. The two knife edges 
are also horizontal, that is, in the plane 
of the axis of the flame tube; each knife 



214 



February 1951 Journal of the SMPTE Vol. 56 




Plane Mirrors 




SIDE VIEW 



Test Section 



Camera 



Knife Edge 




TOP VIEW 
Fig. 1. General ray system for stereoscopic Schlieren. 




Fig. 2. Photo of the system from camera end, showing 6 -ft flame tube 

in position. 

J. H. Hett: Stereoscopic Schlieren System 



215 




Fig. 3a. Stereo pictures of flame front about 25 msec after spark ignition 

of stoichiometric mixture of propane-air in flame tube. 
The heavy black transverse image is a 3-hole grid placed across the tube. 




Fig. 3b. The turbulent flame after passing through the grid. The wire 
mesh was added as a reference frame behind the back surface of the tube. 



216 



February 1951 Journal of the SMPTE Vol. 56 




Fig. 4. The two images of the test hexagon. These can be rotated 

at high speed. 



Non-Polarizing 
Screen 



14 ft. 



Projection Lens, 

60-Mm Effective 

Focal Length 





Dove Prism 



/Deviation Wedges 
/Baffle 




0.800 in. 



Fig. 5. The projection system. The Dove prism is rotated 
until the images are vertical on the screen. 



J. H. Hett: Stereoscopic Schlieren System 



217 



edge cuts the ray bundle from the 
bottom. The illumination is sufficient 
in this sytem to operate the Fastax 
camera at 9000 frames/sec, using Lina- 
graph film. Attempts to use the system 
with Kodachrome film at 500 frames/sec 
were unsuccessful. In adjusting the 
system before exposure it was found 
necessary to view the film directly 
rather than use the regular. Fastax 
viewer, because the relative aperture 
of the viewer is smaller than that of the 
camera lens, thus vignetting some of the 
rays. 

The projector is a Bell & Howell Dip- 
lomat, Model A, modified as follows: 
A two diopter negative lens is com- 
bined with the regular 50-mm projec- 
tion lens to obtain a longer back focal 
length. Between the lens and the film 
gate two wedges are mounted as shown 
in Fig. 5. Also, two crossed polaroids 
are mounted in front of the wedges and 
a baffle is placed to isolate each field. 
A Dove prism is placed in front of the 
projection lens at such an angle that 
the images of the flame tube appear ver- 
tical on the screen. This is necessary 
since the observer can see only stereo- 
scopically in the horizontal plane 
through the eyes. A throw distance of 
14 ft is used to a nonpolarizing screen. 
The observer wears crossed polaroid 
lenses. 

For successful operation of this 
system, the following points should be 
carefully considered: 

1. Screen registration : This involves 
careful setting of the images in the 
Fastax, particularly in the horizontal 
direction (vertical direction on the 



screen). It is desirable to use fine 
wires in the object field and set them 
to within 0.002 in. in the image field. 
At the projector, the deviation angle 
of the prisms, focal length of projec- 
tion lens and throw distance are all 
related quantities and should be care- 
fully adjusted, 

2. Care should be taken to avoid 
image jump on the screen. Even though 
the images may jump together, this 
causes difficulty in stereoscopic fusing. 

For more precise analysis, prints may 
be made from the film and viewed with a 
conventional Wheatstone stereoscope 
using magnification. 

Observational Results 

The system was tested by using 
opaque objects such as the hexagonal 
cylinder which was rotated at high 
speed. Completely satisfactory pic- 
tures were obtained. 

Some loss of the depth effect is en- 
countered in viewing symmetrical 
smooth flame surfaces, as for example, 
the spherical wave leaving the spark 
ignition point of Fig. 3a. This effect is 
probably due to high transparency and 
lack of contrast. ; 

In turbulent flames or turbulent hot 
gases, the stereoscopic effect is achieved. 
If one knows the true scale of the ob- 
ject field, it is possible to locate events 
along the line of sight, and also deter- 
mine directions of rotation of the gases. 

The author wishes to acknowledge the 
assistance of R. J. Kraushaar and S. 
Braunstein with much of the construc- 
tion and experimental work. 



218 



February 1951 Journal of the SMPTE Vol. 56 



Some Commercial Aspects of a New 16-Mm 
Intermediate Film Television System 

By Raymond L. Garman and Blair Foulds 



Theater television requires picture quality comparable to that attained 
in feature film releases, and flexibility in program scheduling comparable 
to television broadcasting. A new 16-mm intermediate film system de- 
signed for these requirements is described. The system includes video 
recording equipment for pickup of coaxial-line or broadcast programs, 
high-intensity film projection equipment, and an automatic rapid film 
processor. The use of the rapid film processor is discussed in connection 
with delay techniques for adequate program scheduling. General operat- 
ing characteristics are analyzed in terms of economics of the system. 



INTEREST in theater television has 
been growing for a number of years. 
The past year has seen considerable 
activity in equipment development 
along two main lines. One has been the 
development of 35-mm film delay 
methods; the other has been the de- 
velopment of better cathode-ray tubes 
for use with Schmidt optical systems 
in direct projection. 

The 35-mm film delay system has 
been recognized from the beginning as 
one which is capable of providing high 
quality performance. The cost of in- 
stallation and operation of such a sys- 
tem is understandably high. 

The direct projection system offers a 
possible operating cost advantage but 
does not, at present, meet all neces- 

Presented on April 25, 1950, at the So- 
ciety's Convention at Chicago, 111., by 
Raymond L. Garman and Blair Foulds, 
General Precision Laboratory, Inc., Pleas- 
antville,.N. Y. 



sary performance requirements. It 
provides only a marginal amount of 
light. Additionally, operation of the 
cathode-ray tube at or near maximum 
light level is usually accompanied by 
poor definition in the highlights. The 
extremely high voltage required by the 
cathode-ray tube (50 to 80 kv in 1949 
practice, with higher voltages in pros- 
pect) has been difficult to contain within 
the equipment and can be expected to 
result in some erratic component fail- 
ures unless extreme precautions are 
taken. 

A third system is under development 
in Switzerland. The details of this 
system, known as the AFIF Television 
Projection System, 1 were reported to 
this Society at its last convention. The 
system is intermediate in principle be- 
tween the delay or film recording sys- 
tem and the direct system. It com- 
bines some of the features of both sys- 
tems. Whether or not the system of- 



February 1951 Journal of the SMPTE Vol. 56 



219 



fers promise for the future remains to 
be seen. 

While these developments have been 
in progress, a new system has been 
quietly reaching maturity. This system 
stems from the 16-mm video recording 
field, which was an infant in the profes- 
sional sense a little more than a year ago, 
and now uses more 16-mm film than all 
other professional and amateur services 
combined. The quality of the film 
produced by video recording has been 
steadily improved so that today it can 
be said that if professional equipment is 
used throughout, from program recep- 
tion to film projection, theater quality 
can be attained. The 16-mm film 
size provides some noteworthy advan- 
tages. Cost is considerably reduced over 
comparable 35-mm systems, and the 
final product is more easily handled, 
shipped and stored. A 16-mm video 
recording system similar to that used 
in television studios can be engineered 
for theater service, thus providing an 
intermediate-film television system. To 
be successful, however, such a system 
must be designed for the specific com- 
mercial requirements of theater pres- 
entation. 

System Considerations 

Commercial use in theaters calls for 
quality equipment throughout. Con- 
version from the television standard 
rate of 30 frames/sec to the 24-frame/sec 
rate of the motion picture industry 
must be performed by a thoroughly 
reliable method. Particular care is 
required in design of the many camera 
and projector details in order to realize 
fully the resolution capabilities of 16- 
mm film. Operation must be almost en- 
tirely automatic, not only to keep the 
amount and types of skilled labor to a 
minimum but also to assure a steady 
flow of high quality product. Com- 
ponents must be as compact as possible 
to permit installation within the limited 
space available in our present day mo- 
tion picture houses, without occa- 



sioning extremely high installation costs. 
Splicing and editing facilities are needed 
to allow for insertion of special trailers, 
titles and show announcements, and to 
contribute generally to the over-all as- 
pect of showmanship. Processing units 
must be designed so as to require the 
minimum amount of chemicals in stor- 
age, because of the frequent shutdowns 
which may be required for program 
scheduling. The projector itself must 
be capable of providing adequate light 
for even the largest motion picture 
houses and of fully utilizing the per- 
formance possibilities of correctly re- 
corded and processed 16-mm sound 
track. Maintenance cost must be low, 
and complicated maintenance opera- 
tions must be avoided wherever pos- 
sible. In order to meet these severe 
commercial requirements, new concepts 
have to be introduced in many or all 
phases of the art. 

Physical Requirements 

A better insight into some of the 
problems can be gained by reviewing 
the physical requirements of the various 
units which comprise a complete theater 
installation. A typical installation, 
shown in skeleton form in Fig. 1, has 
terminal facilities for either off-the-air 
or coaxial-line pickup. The video por- 
tion of the program is photographed on 
16-mm film from the face of the re- 
cording cathode-ray tube, and the 
accompanying sound is recorded simul- 
taneously on the film sound track. Ex- 
posed film from the camera feeds into 
the rapid film processor and emerges as a 
dry, waxed print, ready for projection 
in the special new 16-mm projector de- 
signed for the purpose. 

The antenna array is the starting 
point of the system. For theater use, 
the antenna must provide noise-free 
operation, freedom from ghosts, and 
extremely steady signals. This can be 
achieved by careful selection of the 
antenna for reception from a particular 
station or by the use of multiple an- 



February 1951 Journal of the SMPTE Vol. 56 



ANTENNA VIDEO 

V7 LINE 

X/ INPUT 




Fig. 1. Block diagram of a 16- mm intermediate film theater television system. 



tennas or rotatable antennas if several 
stations are to be received under severe 
conditions. 

The receiver must be reliable and 
stable in operation, easy .to tune, and 
simple to maintain. A receiver which 
provides the necessary reliability and 
stability has been developed, and it has 
been described elsewhere in this JOUR- 
NAL.* By use of remote controls, the 
receiver unit can be placed in some rela- 
tively inaccessible part of the projec- 
tion booth, or even in some other part of 
the theater. 

Programs can be taken from the re- 
ceiver or from separate video and sound 
line inputs. In either case, the video 
portion of the program feeds into a 
block which we choose to call the "Video 
Function Generator." This block con- 
sists entirely of electronic circuit ele- 
ments which perform the various tasks 
required for conversion of a composite 
video signal into a picture suitable for 



photography. It performs the frame- 
rate conversion from television stand- 
ard rate of 30 frames/sec to the motion 
picture standard rate of 24 frames/sec. 
It provides a gray-range, or gamma 
correction, and contains control and 
monitoring facilities. This block can, 
in fact, be regarded as the heart of the 
system. 

Several blocks, including the video 
function generator, the 30-kv high- 
voltage power supply, the recording 
cathode-ray tube, the camera and the 
sound circuits are all contained in a 
single Video Recorder Unit. 3 The unit, 
shown in Fig. 2, is slightly more than 
5 ft wide, and stands 6 ft high to the 
top of the reel housing. It should be 
mentioned that, although a double reel 
housing is shown, only the feed side is 
used. The reel housing has a 1200-ft 
capacity, which is equivalent to 33 min 
of continuous running time. A larger 
reel housing which holds as much as 



Garman and Foulds: Intermediate Film TV System 



221 




Fig. 2. Video recorder. 



4000 ft of film can be installed in place 
of the 1200-ft one. The hood at the 
left conceals the recording cathode-ray 
tube. The electronic circuits associ- 
ated with the video function generator 
are contained in the base. 

The gray-scale or gamma correction 
circuit enables a considerable improve- 
ment in picture quality over that ob- 
tained in an uncorrected system. A 
gamma of between 1.5 and 1.7 is gen- 
erally considered to give the most pleas- 
ing picture for motion picture exhibi- 
tion by direct projection. The gamma 
value actually obtained depends on the 
over-all transfer characteristic of the 
system elements which intervene be- 
tween the scene and the print. Any 
system element having a nonlinear 



transfer characteristic affects the gamma 
value. Amplifying and detecting ele- 
ments which are commonly used have 
linear transfer characteristics, but many 
of the currently available light-to- 
signal and signal-to-light transducers 
are inherently nonlinear. Picture con- 
trast in present television practice ap- 
proaches that of a high-gamma print, 
due to the nonlinear transfer char- 
acteristics of the camera pickup tube 
and the recording cathode-ray tube. 
The film itself further increases the 
final gamma value so that the cumula- 
tive effect in an uncorrected system is a 
higher gamma value than desired. 
The correction circuit, a nonlinear 
amplifier with an adjustable character- 
istic of the type required to reduce print 



February 1951 Journal of the SMPTE Vol. 56 




SECONDS 



Fig. 3. Film path. 



gamma, can be controlled as required 
to produce an optimum print. 

The remaining portion of the video 
function generator is devoted to the 
frame rate conversion circuits. The 
operation of these circuits is based on 
the fact that each television frame 
contains exactly 525 horizontal scan- 
ning lines. Electronic counting circuits 
are therefore used to time the film ex- 
posure. The operation is as follows: 

Film exposure may start at any hori- 
zontal scanning line of the television 
image. Once started, the exposure con- 
tinues until exactly 525 lines have 
been counted out. The circuits then 
blank the recording cathode-ray tube 
and stop the exposure. 

In the camera, film pulldown starts 
after exposure stops. At the end of ^ 4 
sec, both the exposure and film pull- 
down have been completed. The cam- 
era then delivers a cycling pulse which 
starts a new cycle. Photography is 
thus performed at a rate of 24 frames/ 



sec, that is, at the rate established by 
the film camera. A complete televi- 
sion frame is photographed for each film 
frame, even at rates considerably below 

24 frames if required. 

The camera has a sufficiently fast 
pulldown to operate within ^20 sec, 
which is the time interval between 
the end of the television frame and the 
start of the next film frame exposure. 
On the other hand, the camera need 
not have the usual mechanical shutter. 
The counter circuits, in effect, form an 
electronic shutter which has a much 
greater timing accuracy than a me- 
chanical shutter. 

The recording cathode-ray tube pre- 
sents a negative picture to the camera. 
Reversal introduced by subsequent 
processing results in a positive print. 
The cathode-ray tube operates at about 

25 kv, which is considerably less than 
that required in the direct projection 
system. This voltage must, however, 
be closely maintained. In other words, 



Carman and Foulds: Intermediate Film TV System 



223 




Fig. 4. Rapid film processor. 



a "stiff" power supply is needed. 
Otherwise, the change in beam current 
due to differences in average scene 
brightness would cause corresponding 
changes in both picture size and focus 
adjustments. 

The audio side of the system can and 
should achieve better than average 
sound quality. Since playback is 
from the same film original as that on 
which the recording is made, only a 
single recording and playback opera- 
tion is required. Further, the entire 
process can be controlled within the 
theater. If the recording is good, the 
reproduction can be correspondingly 
good. 

The sound recording head in the 
system described is one which was 



developed by J. A. Maurer, Inc. Re- 
sponse is essentially flat to 9 kc. Re- 
cordings are high-level variable-den- 
sity, corrected for both the toe and 
shoulder of the H&D curve. An inter- 
modulation figure of 6% is obtained, 
which is comparable with that of the 
best double-film toe recordings. A 
network matches the recording char- 
acteristic to the playback characteristic 
of the reproducer. The G.P.L. pro- 
jector sound head is designed to match 
the frequency range of the recording 
head. 

The problem of handling the ex- 
posed film after it leaves the camera 
introduces several requirements. Splic- 
ing facilities are needed to allow in- 
sertion of fill-ins, trailers, titles, and 



224 



February 1951 Journal of the SMPTE Vol. 56 



processing leader, and to permit camera 
and projector changeover. Film stor- 
age facilities are needed to allow any 
one of the units to be stopped sepa- 
rately while the other units are run- 
ning. These features are shown in 
more detailed form in Fig. 3. 

Two types of film storage racks are 
available. One, the larger, has a maxi- 
mum delay capacity of about 3 min. 
The other, the smaller, has a maximum 
delay capacity of about 10 sec and 
assembles directly to the side of the 
Rapid Film Processor. 4 The storage 
loop in the smaller rack is sufficiently 
long to avoid possible film breakage 
when adjacent units are simultaneously 
started or stopped. The larger rack 
provides sufficient delay for most splic- 
ing and editing operations. Racks can 
be combined for additional delay, if 
necessary. 

A particular advantage of 16-mm 
film is that it permits the use of rela- 
tively compact equipment for film 
handling and processing. The Rapid 
Film Processor is only 5 ft high and 
approximately 3 ft wide (Fig. 4). The 
larger film storage rack, which contains 
sufficient footage for 3 min. of running 
time, occupies a space about 1 ft wide 
by 6 ft high. It is reasonable to assume 
that space for these components can be 
found in or adjacent to most projection 
booths. 

The use of 16-mm film in theaters has 
previously been limited by the lack of 
projectors with sufficient light output 
for the purpose. An arc lamp pro- 
jector has therefore been developed 
and made available for theater use. 
The standard 16-mm arc lamp pro- 
jector (Fig. 5) provides 2000 total 
screen lumens when used with an //1. 6 
lens. With special carbons and feed 
methods, it may be possible to obtain 
as much as 3200 total screen lumens. 
The illumination and screen brightness 
figures for both types of carbons are 
shown in Table I. The figures are 
stated on the basis of shutter running 








Fig. 5. 16-Mm arc lamp projector. 

and no film in the gate. The acceptable 
screen brightness by SMPTE standards, 
on this same basis, is 9 to 14 foot- 
Lamberts. It is therefore apparent that 
illumination is adequate for even fairly 
large houses. The illumination figures 
shown may seem surprisingly high for 
the film size used. However, the aver- 
age shutter efficiency of 35-mm pro- 
jectors is in the order of 50%, while the 
shutter efficiency of the particular 16- 
mm projector described is 73%. In 
addition, a larger lens aperture is used 
with this projector than is commonly 
used in 35-mm machines. 

In regard to operating costs, 16-mm 
film offers an appreciable advantage 



Carman and Foulds: Intermediate^Film TV System 



225 



Table I. Arc Lamp Projector Performance 







Standard Arc Lamp 


Special Arc Lamp 






for 16-Mm Projector 


for 16-Mm Projector 






2000 Total Screen Lumens* 


3200 Total Screen Lumens* 


T'Virnw T)ictann<i 


Pictur 






j. nrow J 'isi a IH t , 
in feet 


Width, 




Screen Brightness, 




Screen Brightness, 










TlliiT^inn 




4Tn 


2-In 


in feet 


inuniina- 


lOot-lj&mDsrtQ 


ill urn ins."* 


I O< >t - 1 jiilll I >tM' t .S 


-in. 
F.L. 


F.L.' 




foot- 


Matte 1 Beaded 


tion, 
foot- 


Matte 


Beaded 


Lens 


Lens 




candles 


Screen 


Screen* 


candles 


Screen 


Screen 


80 


40 


7.5 


47 


38 


277 


76 


61 


443 i 


100 


50 


9.5 


30 


24 


175 


48 


39 


281 ' 


120 


60 


11.4 


21 


17 


121 


33 


27 


193 


140 


70 


13.2 


15 


12 


90 


25 


20 


143 


160 


80 


15.2 


11 


9.1 


66 


18 


15 


106 


180 


90 


17.1 


9.3 


7.5 


55 


15 


12 


|87 


200 


100 


19.0 


7.3 


5.9 


43 


12 


9.4 


69 


220 


110 


21.0 


6.1 


4.9 


35 


9.7 


7.8 


57 


240 


120 


22.8 


5.1 


4.1 


30 


8.2 


6.6 


48 


260 


130 


24.6 


4.4 


3.5 


26 


7.0 


5.7 


41 


280 


140 


26.6 


3.8 


3.1 


22 


6.1 


4.9 


36 


300 


150 


28.5 


3.3 


2.6 


19 


5.2 


4.2 


31 


340 


170 


32.4 


2.5 


2.0 


15 


4.1 


3.3 


24 


400 


200 


38.0 


1.9 


1.5 


11 


3.0 


2.4 


17 



* With shutter running, no film in machine. 

J When viewed in the direction of the projected beam. When viewed 10 from the projected beam 
direction, the values are 40% of those listed. 



over 35-mm film. Current film prices 
for fine grain release positives on safety 
stock are $17.20 per hour for 16-mm 
film, as against $80.00 per hour for 
35-mm film. Other operating costs 
can also be expected to be lower, but 
not necessarily in as high a ratio, nor 
can exact evaluations be made. Chemi- 
cal costs should be lower with 16-mm 
film. The greater compactness of 16- 
mm equipment, and the smaller floor 
space requirement, may allow a saving 
in new theater construction cost or, 
alternately, an increase in useful seat- 
ing capacity. The life expectancy of 
16-mm equipment which is designed 
for professional use is as high as, or 
higher than, that of 35-mm equipment. 
The projector operates at the same 
frame rate in either case and can there- 
fore be assumed to have the same life 
expectancy; the processor, operating at 
a lower film travel rate, can be assumed 
to have a greater life expectancy. Due 
to the ease of handling 16-mm film, 
labor costs may be somewhat lower. 



Conclusion 

All of the components for intermedi- 
ate film theater television using 16-mm 
film size are currently available. This 
film size offers important cost econ- 
omies in both hourly operation and 
initial installations. Delay techniques 
permit flexibility in program scheduling. 
Picture quality and screen brightness 
are entirely satisfactory for theater 
use. 

References 

1. "Theater Television," Jour. SMPE, 
vol. 52, pp. 243-272, Mar. 1949. (Ap- 
pendix, pp. 263-266, describes AFIF 
Television Projection System.) 

2. F. N. Gillette and J. S. Ewing, "Com- 
ponent arrangement for a versatile 
television receiver," Jour. SMPTE, 
vol. 55, pp. 189-196, Aug. 1950. 

3. F. Gillette, G. King and R. White, 
"Video program recording," Electronics, 
vol. 23, no. 10, pp. 90-95, Oct. 1950. 

4. J. S. Hall, A. Mayer and G. Maslach, 
"A 16-mm. rapid film processor," 
Jour. SMPTE, vol. 55, pp. 27-36, 
July 1950. 



226 



February 1951 Journal of the SMPTE Vol. 56 



Television Film Recording and Editing 



By Albert Abramson 



This paper reviews the uses of television film recording and the possibilities 
of applying the editing principle to it. 



THERE is A NEED in television for a 
flexibility and perfection that can- 
not be attained by using live television 
techniques. The means for meeting 
this need lie within the scope of any 
television station equipped to record 
television programs on film. But to- 
day's methods of television film record- 
ing must be improved both filmically 
and technically. 

Cathode-ray photography dates back 
to 1938. In that year the first attempts 
were made to photograph the image on 
the kinescope tube. 1 The low light 
intensity of the image combined with 
the use of standard spring-wound 
cameras gave very unsatisfactory re- 
sults. The most difficult problem was 
to synchronize the 30-frame/sec rate of 
the television screen with the 24-frame/ 
sec rate that is standard motion picture 
practice. Twenty-four frames per sec- 
ond were necessary in order to use ex- 
isting projection and sound apparatus. 
The most critical characteristic in the 
recording camera is the timing of the 
shutter blanking and exposure interval. 2 
This problem has been solved by means 

A contribution submitted on May 31, 
1950, by Albert Abramson, Graduate 
Student, Cinema Dept., University of 
Southern California, 3441 W. Second St., 
Los Angeles 4, Calif. 



of cameras incorporating specially de- 
signed mechanical or electronic shutters. 

Essentially, a television film recorder 
consists of this special camera, a monitor 
which will give precise visual images and 
a sound recorder to pick up the accom- 
panying sound. At present there are 
both 16-mm and 35-mm television film 
recorders with either single or double 
system sound. 

Using 16-mm has the advantage of 
lower film and processing costs; it is 
approximately one third as expensive as 
35-mm. No marked improvement is 
to be had by recording on 35-mm rather 
than 16-mm at the present time. With 
the use of fine-grain, high-resolution, 
16-mm film emulsions, no loss of resolu- 
tion in recording the television image is 
noticeable. Using 35-mm has the added 
disadvantage of very stringent fire 
regulations and, finally, the cost of 35- 
mm projection equipment is often 
prohibitive. As a result, most tele- 
vision stations are using 16-mm film for 
their recordings 3 ; 35-mm film is being 
used primarily for theater television. 

There are four main purposes for 
which television film recordings can be 
made at present : 

1. Transcriptions. The transcription 
is the mam function of television film 
recordings today. It is a recording of a 



February 1951 Journal of the SMPTE Vol. 56 



227 



complete show either as it goes over the 
air or as a closed-circuit operation. It 
may be shown as: (a) delayed telecast, 
to make up for the difference in time 
zones between the east and west coasts; 
(b) repeat telecast, to catch a larger 
audience at a more appropriate time; 
or (c) syndicated telecast, in which case 
it is sent to a station that is not con- 
nected by either coaxial cable or micro- 
wave relay, and is shown at any con- 
venient time. 

2. Theater television. Television film 
recordings are used as an intermediate 
system of television projection. The 
program is picked up by receiving equip- 
ment at the theater. It is then re- 
corded by 35-mm single system equip- 
ment. The signal is inverted and a 
direct-positive print results. The film 
is fed into rapid-processing machines 
where it is processed, dried and fed 
directly into the projection machine in a 
little over a minute from the time of ex- 
posure. This system allows theaters to 
show television programs using existing 
35-mm projection equipment. 4 

3. Research. This includes record- 
ings made for either auditions or pre- 
views. Recordings are often made to 
improve the quality of a program. 
Techniques of camera work, acting, 
lighting, set design and all the elements 
that go into a television show can be 
checked before the program is to go on 
the air. 

4. Reference. There is no better way 
to keep a record of a television program 
than to record it as it goes over the air. 
It is possible that the F.C.C. may re- 
quire a record kept of every program 
telecast. 

Technically speaking, the quality of 
television film recording is fairly good 
and will continue to improve. With 
certain refinements, such as greater 
bandwidth and more lines, it should 
eventually be impossible to distinguish 
a television film recording from a film 
shot by a standard motion picture 
camera. 



Dissatisfaction with present television 
film recording quality has led to the 
rise of the multicamera motion picture 
system. In this system a multiple 
camera setup utilizing three or more 
standard motion picture cameras is 
used. All cameras can operate simul- 
taneously. By utilizing live television 
techniques of dollying and camera 
movement, the program is covered from 
a multitude of angles. With the use of 
an ingenious cuing system the films 
from the different cameras are later 
spliced together to form a complete 
television program. 5 The use of multi- 
camera setups is, of course, not new. 
They were extensively used some twenty 
years ago in the early days of sound. 6 
This system, at present, gives major- 
studio quality and as such deserves 
much merit. Assuming that eventually 
the quality of television film recording 
will equal that of standard motion pic- 
ture practice, the multicamera system 
will not maintain its superiority over 
recording through the television camera 
which has these advantages: 

1. The television camera has an 
enormous advantage over the standard 
motion picture camera, in that all it 
"sees" can be viewed instantly. All 
camera setups can be checked on the 
monitor for lighting and composition. 
There is no problem of parallax, focus 
or exposure. The director knows in ad- 
vance exactly what the scene will look 
like. Many a director and cameraman 
in the major film industry would like to 
have this tremendous advantage. Dur- 
ing the actual performance any mistakes 
can be seen and immediately reshot. 
There need be no waiting for "rushes" 
as there can be no doubt as to the 
scene's outcome. 

2. It allows film to take advantage of 
the light amplification characteristics 
of the image orthicon camera. Thus it 
will be possible to film certain scenes 
under light conditions that are im- 
possible with the standard motion pic- 
ture camera. 7 This means the use of 



228 



February 1951 Journal of the SMPTE Vol. 56 



more natural lighting or the use of a 
minimum of lighting equipment. 

3. Certain optical effects such as dis- 
solves, fade-ins, fade-outs, double ex- 
posures and background shots can be 
made in the television cameras them- 
selves. This adds to the economy of the 
system as it can reduce the cost of pro- 
ducing these special effects optically. 

4. It allows the television station to 
utilize the equipment on hand. Thus 
the television camera can serve a dual 
purpose. It can be used for live tele- 
vision programs or it can be used in 
conjunction with the television film re- 
corder. This allows the station com- 
plete control over program content as 
all programs can be made on the studio 
premises and conform to the station's 
needs. 

5. Since all recording is accomplished 
at a central point, it should be easy to 
keep the recording and developing proc- 
ess under the strictest possible control. 
The potentialities of a system like this 
are unlimited and may make the stand- 
ard motion picture camera as we know 
it today obsolete. 

Filmically speaking, the present-day 
television film recording leaves much to 
be desired. It possesses the physical 
characteristics of motion picture film 
but lacks the inherent capabilities of the 
true motion picture, for it is restricted 
by the limitations of the live television 
program. 

In a live television program a unity 
of time and space must be observed. 
Movement is confined by the physical 
limits of the stage itself. Performers 
must learn complete scripts. Changes 
in costume or makeup take time and 
there must be cover-up action during 
this period. Even when using two or 
more sets the performer can travel 
through them only at a certain speed. 
Transitions must either be eliminated or 
filmed in advance. Outdoor sets are 
seldom if ever used. During the per- 
formance any mistake is easily noticed 
and there is no chance to rectify it. 



As a result of these restrictions, the 
average television play today resembles 
a stage play in that the story is ad- 
vanced through the dialog. This is 
good stage technique, but is poor tele- 
vision. Television is a visual medium 
and as such the story could and should 
be advanced by visual means. Move- 
ment on the screen is interesting and 
tells its own story. Dialog is important 
but should never dominate the picture. 

Editing 

The true motion picture is not just a 
recording of reality but a rearrangement 
of that reality to suit its own purposes. 
Both the standard motion picture 
camera and the television film recorder 
are recording mechanisms. They can 
do nothing but record on film a scene 
that is placed before their lenses. 
Then how does the motion picture gain 
its flexibility and freedom of movement, 
its ability to manipulate time and space? 
The answer lies in the editing process. 
It is in the editing room that the motion 
picture, as we know it, comes into being. 

Here is created filmic time and filmic 
space. Filmic time and filmic space 
exist only on the individual strips of 
motion picture film. Actual events can 
be stretched or compressed. Time can 
be made to stand still or to go forward or 
backward. It is possible to show 
events, occurring at widely separated 
points, and simultaneously. Unrelated 
shots are cut together and meaning is 
extracted from their juxtaposition. An 
accident occurs; we see, in rapid suc- 
cession, the victim crossing the street, 
the driver's grim look, his foot slam on 
the brake, the victim's horrified face, 
the wheels skidding on the pavement, 
the victim lying in the street. A man 
steps out of a New York hotel into a 
South Ameri can street . These and many 
other scenes are possible only through 
the editing process. These are no mere 
tricks, they are the lifeblood of the 
visual medium. As a visual medium, 
television can use the editing principle 



Albert Abramson: Use of Kinescope Recorder 



229 



to its advantage. This will free tele- 
vision from the limitations imposed 
upon it by live television techniques. 

At the present time, the major net- 
works are editing television film record- 
ings for the following reasons : 

1. To make transitions which would 
otherwise be impossible if the program 
were recorded straight through. 

2. To rerecord imperfect scenes. 

3. To eliminate excess footage and 
edit the show down to required length. 

In applying the editing principle to 
television film recording, it is well to 
note a major difference between the 
motion picture and television. Both, 
being visual mediums, have the shot as 
their foundation. However, the motion 
picture is filmed on a single-shot basis 
whereas television is set up on a mul- 
tiple-shot basis. This is no handicap; 
quite the contrary, it can be used to 
great advantage. 

In the motion picture each individual 
shot is arranged for maximum effect. 
There is always one certain camera 
angle that will be most effective de- 
pending upon what idea is being con- 
veyed to the audience. Thus each 
scene is carefully arranged to put across 
this idea. Therefore, even the same 
scene when photographed from different 
angles will be rearranged to suit each 
individual shot even though all of these 
shots will be cut together to create a 
seemingly continuous scene. 

This is not necessary in television, for 
the use of multiple cameras combined 
with electronic cutting makes it possible 
to get a variety of shots without making 
new setups for each individual shot. 
This can be done by careful planning of 
camera angles, the use of proper focal 
length lenses and the use of lighting to 
suit the scene. Here, of course, the 
maximum effect from each shot or cut is 
not as assured as in the single-shot setup, 
but such should be nearly attained. 
Thus it is possible to create a maximum 
number of shots with a minimum num- 
ber of setups for any given scene. It is 



proposed to use this type of multi- 
camera setup wherever the action will 
allow it. 

In order to apply the editing prin- 
ciple to television film recording, pre- 
production planning is the first 
necessity. In addition to planning de- 
tails of sets, costumes, props, etc., the 
script must be broken down into two 
types of sequences. The first type of 
sequence should consist of that kind of 
scene where two or more television 
cameras can be used for the necessary 
variety of shots. This type of scene will 
be recorded as a unit making full use of 
electronic cutting. 

The second type of sequence should 
consist of that kind of scene where it is 
necessary to stop the recorder to make 
changes in lighting, costumes, sets, 
makeup, etc. This can be recorded 
with a multi-camera setup or, if cir- 
cumstances demand, with only one 
camera. 

In all instances, the various sequences 
will be recorded in whatever order is 
most practical. By minute scheduling 
of operations it should be possible to 
record the various sequences in the 
shortest amount of time. After proc- 
essing, the recorded sequences can then 
be edited into a smooth, flexible program 
with a minimum of time and effort. 
The cost should approximate straight 
television film recording with the quality 
equaling that of professional motion pic- 
ture practice. 

This is a process in which we are 
utilizing the best features of motion 
pictures and television. We have given 
the television camera a memory. We 
have taken the unique picture-control 
elements of the television camera and 
added the permanency, flexibility and 
perfection of motion picture film. Thus 
the process is one that is peculiar to 
neither motion pictures nor television 
alone, but is a synthesis of the two that 
can be used to their mutual advantage. 

It has been said that television will 
lose its sense of "immediacy" through 



230 



February 1951 Journal of the SMPTE Vol. 56 



the use of television film recordings. 
It has also been said that the public 
likes to know that the program being 
telecast is being presented at that very 
moment. This all depends upon the 
type of program being considered. 
Every day millions of persons attend 
motion pictures that are from nine to 
twelve months old before being released. 
Even the earliest 'newsreels' are from a 
few days to several months old when 
being presented. 

"Immediacy" is determined by pro- 
gram content. Obviously, no other 
medium is as well equipped to present 
an event as it actually happens as tele- 
vision. In reporting spot news, sport 
events, presidential elections and other 
events of great public interest, television 
can present these programs to the public 
as they actually occur. To this list 
can be added programs of a semire- 
hearsed nature such as vaudeville, 
comedy or variety shows. However, 
rehearsed programs, especially drama of 
all varieties, need the perfection and 
flexibility that only a filmic use of the 
television film recorder can give them. 
There is no reason why drama should be 
presented live. The perfection and 
flexibility that can be had by this 



method mean better dramatic programs 
and that is our ultimate goal. 

This is where the future of television 
film recording lies and it doesn't inter- 
fere with any type of program in which 
"immediacy" is its most important 
aspect. To the contrary, it can record 
those programs of lasting interest and 
preserve them for posterity. 

References 

1. R. M. Fraser, "Motion picture photog- 
raphy of television images," RCA 
Rev., vol. 9, pp. 202-217, June 1948. 

2. Thomas T. Goldsmith, Jr. and Harry 
C. Milholland, "Television transcrip- 
tion," Electronics, vol. 21, pp. 68-71, 
Oct. 1948. 

3. Report of the Television Committee, 
"Films in television," Jour. SMPE, 
vol. 52, pp. 363-383, Apr. 1949. 

4. Richard Hodgson, "Theatre tele- 
vision system," Jour. SMPE, vol. 52, 
pp. 540-548, May 1949. 

5. Jerry Fairbanks, "Motion picture pro- 
duction for television," Jour. SMPTE, 
vol. 55, pp. 567-575, Dec. 1950. 

6. F. Green, The Film Finds its Tongue, 
p. 67, G. P. Putnam, New York, 1929. 

7. Thomas T. Goldsmith, Jr. and Harry 
C. Milholland, "Television transcrip- 
tion by motion picture film," Jour. 
SMPE, vol. 51, pp. 107-117, Aug. 
1948. 



Albert A brain. son: Use of Kinescope Recorder 



231 



ABSTRACT 



The Genlock-A New Tool for Better TV Programming 



By John H. Roe 



RECENTLY, the need for more and 
better techniques in video pro- 
gramming has become more and more 
apparent, particularly as picture qual- 
ity has improved, thus focusing atten- 
tion on ideas for adding some of the 
finer touches. One of the gaps in the 
present programming structure arises 
from the lack of synchronization be- 
tween two distinct program sources 
which may supply successive parts of a 
program. The field-frequency pulses 
may be phased together by manual 
adjustment and they will stay so as 
long as the same power source is the 
reference for both generators, but there 
is no such simple solution to the prob- 
lem of phasing the line-frequency pulses. 
Lack of tight lock-in between two 
such systems results in several program- 
ming limitations. For example, when 
the program line is switched from one 
system to the other, the receivers have 
to adjust themselves to the new syn- 
chronizing signal. The horizontal (line- 
frequency) scanning changes very 
quickly in most cases, but the vertical 
(field-frequency) scanning circuits have 
much more inertia and do not respond 
quickly. The usual result is, therefore, 

Abstract by Pierre Mertz of a paper pre- 
sented on September 26, 1950, at the Na- 
tional Electronics Conference at Chicago, 
111. (in which the SMPTE Central Section 
participated), by John H. Roe, Radio 
Corporation of America, RCA Victor 
Div., Camden, N.J. The complete paper 
will be published in Proceedings of the 
National Electronics Conference, vol. 6 
(for 1950). 



that the picture on a receiver will "roll 
over," much to the annoyance of the 
viewer. 

Another limitation is the impossibil- 
ity of using lap-dissolves and super- 
positions involving pictures from two 
unrelated television pickup systems. 
The increasing use of lap-dissolves and 
superpositions in studio programs makes 
it seem more and more desirable to 
provide means to produce the same 
effects at all times regardless of the 
sources of the signals to be treated. To 
make them possible, the synchronizing 
signal generators must be locked to- 
gether tightly, field for field and line for 
line, just as though the whole system 
were operating on one generator instead 
of two. 

The most direct solution to this prob- 
lem is to provide means for locking the 
local synchronizing signal generator, 
as a slave, to the remote generator, as a 
master. Once the equipment for this 
control of the local generator is func- 
tioning, the remote signals may be 
treated as local signals in any of the 
common types of switching transitions 
and superpositions, thus making it 
possible to go back and forth from one 
source to the other without concern as 
to the point of origination. 

Foreseeing the need and the demand 
for simple, automatic and foolproof 
means for tying two television pickup 
systems together, RCA engineers have 
developed a device called the Gen- 
lock, which accomplishes the desired 
lock-in automatically without any man- 
ual phasing adjustment whatever. 



232 



February 1951 Journal of the SMPTE Vol. 56 



The Genlock 

The Genlock is a unit which com- 
bines two separate circuits which serve 
to provide control signals to the line- 
frequency and field-frequency sections, 
respectively, of the local synchronizing 
signal generator. 

The first consists of an automatic fre- 
quency-control discriminator which de- 
rives a varying d-c error signal from the 
comparison of the horizontal driving 
signal (from the local synchronizing 
signal generator) with the separated 
synchronizing signal derived from the 
remote picture signal. This latter 
synchronizing signal must be separated 
from the composite picture signal in 
some other equipment such as the RCA 
TA-5C stabilizing amplifier. No sepa- 
rator circuit is provided in the Genlock. 
The error signal is applied to the re- 
actance tube in the local synchronizing 
signal generator, thus directly control- 
ling the frequency and phase of the 
master oscillator. The control is rigid, 
allowing no perceptible horizontal drift 
or instability between the two pic- 
tures. 

The second circuit compares the syn- 
chronizing signals, one from the local 
synchronizing signal generator and the 
other from the synchronizing signal 
separator operating on the remote pic- 
ture signal, and from this comparison 
derives an error signal in the form of a 
positive pulse recurring at field fre- 
quency. As long as the two field-fre- 
quency signals are out of phase, the 
pulse exists, but as soon as they become 
coincident, the error pulse ceases to 
exist. The error signal is applied to the 
7:1 counter circuit in the local syn- 
chronizing signal generator (RCA TG- 
1A or TG-10A) in such a way as to 
cause it to miscount. As long as the 
error signal continues to recur, the 
local field frequency drifts at an ac- 
celerated pace causing the two signals 
to approach in phase. At the instance 
of coincidence the error signal dis- 



appears and the counter circuit begins 
to operate normally. Thereafter the 
two signals remain in phase as long as 
the Genlock continues to function. 

The operation of the line-frequency 
control circuit is quite rapid so that 
lock-in of the horizontal scanning cir- 
cuits appears to be almost instantane- 
ous. The field-frequency control cir- 
cuit, however, requires a variable 
amount of time to assume full control 
depending on the initial phase differ- 
ence between the two signals. Phase 
shift brought about by the control 
occurs at a definite rate of three scan- 
ning lines per field. The maximum 
time required to achieve control is 
about 1.46 sec. 

The Genlock never requires more 
than one field to bring the field-fre- 
quency pulses into phase. The reason 
is that when it causes the counter in 
the synchronizing signal generator to 
miscount, it is possible, under the 
proper conditions, to bring about a con- 
version of an "even" to an "odd" field, 
or vice versa. 

The question may arise as to what 
happens if by some mischance the even 
field in one system is brought into 
coincidence with the odd field of the 
other system. The answer is that 
nothing serious takes place. The tops 
and bottoms of the two pictures are 
slightly displaced under such condi- 
tions. 

From a practical point of view, it is 
not important to have exact coincidence 
of the top and bottom lines, respec- 
tively, in the two picture signals. 
Any lack of coincidence means simply 
that the edges of the two vertical blank- 
ing signals are slightly separated in 
time, and therefore, in space, on the 
picture tube. This results in a shift 
up or down, of the top and bottom of 
the raster at the time of switching by 
an amount proportional to the dis- 
crepancy. If the discrepancy is, for 
example, only one or two half-line 
intervals, the shift is almost impercep- 



John H. Roe: The Genlock (Abstract) 



233 



tible. In the average receiver it is hid- 
den behind the mask and is not visible 
at all. 

Thus it may be seen that the Genlock 
is entirely automatic in operation, and 
requires only the proper information 
in the form of suitable signals to bring 
about a solid "marriage" of the two 
synchronizing signal systems. The 
only control necessary is a switch for 
disconnecting the normal frequency 
reference standard and at the same 
time connecting the output of the 
Genlock to the proper circuits in the 
local synchronizing signal generator. 

System Considerations 

Two methods of using the Genlock 
in a television station are suggested. 
In the first case, where only one syn- 
chronizing signal generator is available 
at the studio, the connections between 
it and the Genlock are made through a 
switch. In the second case, where an 
additional or standby synchronizing 
signal generator is available at the 
studio, the Genlock is used to control 
the standby generator, and the system 
is transferred to Genlock operation by 
switching from one generator to the 
other. This is the preferable method 
because it permits previewing of Gen- 



lock operation before the system is 
transferred. 

In either case, because a transfer in 
or out of Genlock operation causes a 
transient disturbance in the operation 
of deflection circuits in receivers, it is 
desirable to make the transfer with the 
video output of the station faded down 
to black. 

Inasmuch as the Genlock makes the 
local station dependent on a signal 
source which is remote and beyond the 
control of local operators, it is inter- 
esting to know what happens when the 
remote signal fails. The Genlock is so 
designed that, when the remote signal 
is lost, the local synchronizing signal 
generator continues to operate quite nor- 
mally at a rate which is very close to that 
existing under Genlock control. In 
other words, the synchronizing signal 
generator becomes free-running, depend- 
ing only on the stability of its master 
oscillator. When the remote signal is re- 
stored, the Genlock regains control in the 
same way as when initially put into oper- 
ation. 

Acknowledgment: Credit is due to 
F. W. Millspaugh and A. H. Turner, 
who contributed much to the develop- 
ment of this device, and to Dr. H. N. 
Kozanowski, under whose direction the 
work was done. 



284 



February 1951 Journal of the SMI* TE Vol. 56 



Standards 



Standards Symbol Changed to PH 



IT HAS BEEN ASA's practice to designate 
each related group of its various activi- 
ties with a single letter symbol. The 
letter "Z" was reserved for those "mis- 
cellaneous" committees not considered 
large enough to warrant assignment of a 
separate symbol. The 30-year old Sec- 
tion Committee on Photography, Z38, 
has been in that category along with 
Sectional Committee on Motion Pictures 
Z22. 

Phenomenal growth of interest in 
photographic standards in recent years 
has so expanded Z38 that it became too 
large to function efficiently under its old 
organization, so ASA and the committee 
members agreed to certain essential 
changes. Z38 was divided into four 
separate committees, which together 
with Z22 were then placed under the 
administration of a newly established 
Photographic Standards (Correlating) 



Committee. A new letter designation 
was established to cover the entire 
group. In view of the imminent need for 
a double-letter code system, the letter 
P, first proposed because it had not been 
used before, was expanded to PH and is 
now the common symbol for all sectional 
committees on photography. ASA has 
officially assigned the following designa- 
tions to the following committees : 
PHI Films, Plates and Papers 
PH2 Photographic Sensitometry 
PH3 Photographic Apparatus 
PH4 Photographic Processing 
PH22 Motion Pictures 
These changes in no way affect the 
scope or membership of the Sectional 
Committee on Motion Pictures, but 
change only the code numbers on all 
new standards. The first proposed 
standards to carry the new numbers 
follow in this issue. 



Cutting and Perforating 32-Mm Film 



Two APPROVED American standards for 
cutting and perforating 32-mm film 
appear on the following pages. These 
standards were first published in the 
February, 1949, JOURNAL as proposals 
to elicit comments or criticisms. Since 
no adverse criticism was received, they 
were processed through the required 
channels and officially approved on 
October 6, 1950. 

Although film of this type has been 
used commercially since 1934, there 
never has been a formal standard. Dur- 
ing the intervening years a number of 



changes have been made in the dimen- 
sions. Debrie, who was the originator 
of the slit-film process for release print- 
ing, was aware that slitting of 32-mm 
film into two 16-mm widths might be 
inaccurate. This inaccuracy would 
make one half wider than the other half, 
and could cause trouble in the projector 
gate. Therefore, he made the original 
French film narrower than twice the 
width of 16-mm film. The first French 
film was about 1.252 in. in width. 
Manufacturers in this country made 
film of this width for some time but later 



February 1951 Journal of the SMPTE Vol. 56 



235 



widened it by 0.005 in. to make it 
1.257 in. 

It appears that there have been four 
or five slightly different styles of per- 
forating in use at various times. Values 
currently adopted for film width and for 
transverse pitch of the perforations are 
believed to be acceptable to all manu- 
facturers. Differences between the 



present standards and the earlier dimen- 
sions are so slight, it is doubtful that 
the users can perceive them. Dimen- 
sions of the perforation, longitudinal 
pitch, and the like, are the same as those 
of current 16-mm film. Dimensioning 
of the drawings is consistent with the 
standards for 16-mm raw stock (Z22.5- 
1947 and Z22.12-1947). 



16-Mm Projection Reels 

PUBLICATION in the February, 1950, 
JOURNAL of a proposed complete revi- 
sion of the American Standard for 16- 
Mm Projection Reels, Z22. 11-1941, 
resulted in a number of comments. 
Consideration of these comments by the 
16-Mm and 8-Mm Motion Pictures 
Committee, which developed this pro- 
posal, has led to recommendation of 
changes in dimensions R, S, and T, and 
in Note 7 (formerly Note 3). Although 



not apparent on the surface, the in- 
tent of these changes is to make possible 
the design of plastic reels within the 
standard dimensions. Because of the 
nature of these changes, the Standards 
Committee agreed that the revised 
proposal, as it appears on the following 
pages, should be republished for ninety 
days trial and criticism. Please send 
comments to Henry Kogel, Staff Engi- 
neer at Society Headquarters, before 
June 1, 1951. 



Projection Lamps 

PROPOSED STANDARDS for two types of 
projection lamps, developed by the 16- 
Mm and 8-Mm Motion Pictures Com- 
mittee, appear on the following pages. 
They are published here for trial and 
criticism for a period of ninety days. 
Please forward any comments to Henry 
Kogel, Staff Engineer at Society head- 
quarters, by June 1, 1951. 

The first of these two proposals, PH 
22.84, is entitled Dimensions for Pro- 
jection Lamps, Medium Prefocus Ring 
Double-Contact Base-Up Type for 16- 
Mm and 8-Mm Motion Picture Projec- 
tors. It shows a type of base developed 
recently to provide improved filament 
positioning, better cooling, and easier 
replacement with the objective of mak- 
ing the lamp compatible with other re- 
cently refined projector elements. In a 
way, this design has been an ideal sub- 
ject for standardization in that it is not 



yet in widespread use and consequently, 
once the general scheme was agreed 
upon, the Committee did not have to 
compromise because of existing practices. 
The base covered by the proposal is the 
subject of a patent assigned to the Gen- 
eral Electric Company. However, after 
a search by the Society disclosed no 
other active patents on pertinent bases, 
rings or sockets, the General Electric 
Company agreed to dedicate this patent 
to the public, clearing the way for 
standardization. 

The second proposed standard is for 
Dimensions for Projection Lamps, Me- 
dium Prefocus Base-Down Type for 16- 
Mm and 8-Mm Motion Picture Projec- 
tors, PH22.85. It will be recognized 
that lamps of this design have been in 
general use for many years; however, 
there has been no American Standard for 
the dimensions. 



236 



February 1951 Journal of the SMPTE Vol. 56 



American Standard 

Cuffing and Perforating Dimensions for 
32-Millimeter Sound Motion Picture 
Negative and Positive Raw Stock 


ASA 

Kre- U. S. Pat. Off. 

PH22.71 -1950 
(Z22.71 - 1950) 


*UDC 778.534.4 











Page 1 of 2 Pages 

] 










D 

D^ 




F 


* 
T 

k 

n 


i 


Cf 

jri 

n 




r 










Dimensions 


Inches 


Millimeters 


A 
B* 
C 
D 
E 
G 

1 
L ** 

R 


1.257 0.001 
0.300 0.0005 
0.0720 0.0004 
0.0500 0.0004 
0.036 0.002 
Not > 0.001 
1.041 0.002 
30.00 0.03 
0.010 0.001 


31.93 0.025 
7.620 0.013 
1.83 0.01 
1.27 db 0.01 
0.91 0.05 
Not > 0.025 
26.44 0.05 
762.00 0.76 
0.25 0.025 


These dimensions and tolerances apply to the material immediately after 
cutting and perforating. 
* In any group of four consecutive perforations, the maximum difference of 
pitch shall not exceed 0.001 inch and should be as much smaller as possible. 
** This dimension represents the length of any 100 consecutive perforation 
intervals. 


Approved October 6, 1950, by the American Standards Association, 
Sponsor: Society of Motion Picture and Television Engineers 


Incorporated 

Universal Decimal Classification 



Copyright, 1950, by American Standards Association, Inc.; reprinted by permission of the copyright holder. 



February 195!i Journal of the SMPTE Vol. 56 



American Standard 

Cutting and Perforating Dimensions for 
32-Millimeter Sound Motion Picture 

M A J D M. D Cx I 

Negative and Positive Raw Stock 



gtii-it 

(Z22.71 - 1950) 



Page 2 of 2 Page* 



Appendix 

The dimensions given in this standard represent the practice of film manu- 
facturers in that the dimensions and tolerances are for film immediately after 
perforation. The punches and dies themselves are made to tolerances con- 
siderably smaller than those given, but owing to the fact that film is a plastic 
material, the dimensions of the slit and perforated film never agree exactly 
with the dimensions of the punches and dies. Shrinkage of the film, due to 
change in moisture content or loss of residual solvents, invariably results in 
a change in these dimensions during the life of the film. This change is 
generally uniform throughout the roll. 

The uniformity of perforation is one of the most important of the variables 
affecting steadiness of projection. 

Variations in pitch from roll to roll are of little significance compared to 
variations from one sprocket hole to the next. Actually, it is the maximum 
variation from one sprocket hole to the next within any small group that is 
important. This is one of the reasons for the method of specifying uniformity 
in dimension B. 

Thirty-two-millimeter release print stock is slit, after printing and develop- 
ing, to 16- mm width. Since a possible error is involved in this slitting, the 
width of 32-mm film is made 0.001 inch narrower than twice the width of 
standard 16-mm film. This narrowing gives a tolerance of 0.001 inch in this 
secondary slitting operation. If the error of slitting exceeds this tolerance, 
one of the 16-mm halves may exceed the width allowed for 16-mm film 
and cause interference in the gate of a projector. In addition to errors of 
centering, there are errors caused by recurring variations in width. These 
errors will cause weave on the screen even though the maximum width of 
the film may not be great enough to cause interference in the projector gate. 



February 1951 Journal of the SMPTE Vol. 56 



American Standard 

Cutting and Perforating Dimensions for 
32-Millimeter Silent Motion Picture 
Negative and Positive Raw Stock 


A5A 

Kre. V. S. Pat. OH. 

PH22.72-1950 

(Z22.72-1950) 


*UDC 778.5 


- 








"I 

JL 


Pag* 1 of 2 Pages 


a 


^ a 


D 


a 

EL 


n a 1_ 


D" 


"7 

g,e 


i__ 

F 

y 


E^- 


p D 


,3 








Dimensions 


Inches 


Millimeters 




A 
B* 
C 
D 
E 
G 

1' 
J 
L** 
R 


1.257 0.001 
0.300 0.0005 
0.0720 0.0004 
0.0500 0.0004 
0.036 0.002 
Not > 0.001 
1.041 0.002 
0.413 0.001 
0.071 0.001 
30.00 0.03 
0.010 0.001 


31.93 
7.620 
1.83 
1.27 
0.91 
Not 
26.44 
10.490 
1.803 
762.00 
0.25 







> 

rb 





0.025 
0.013 
0.01 
0.01 
0.05 
0.025 
0.05 
0.025 
0.025 
0.76 
0.025 


These dimensions and tolerances apply to the material immediately after 
cutting and perforating. 
* In any group of four consecutive perforations, the maximum difference of 
pitch shall not exceed 0.001 inch and should be as much smaller as possible. 
** This dimension represents the length of any 100 consecutive perforation 
intervals. 


Approved October 6, 1950, by the American Standards Association, 
Sponsor: Society of Motion Picture and Television Engineers 


Incorporated 













Copyright, 1950, by American Standards Association, Inc.; reprinted by permission of the copyright holder. 



February 1951 Journal of the SMPTE Vol.56 



239 



American Standard 

Cutting and Perforating Dimensions for 

32-Millimeter Silent Motion Picture 

Negative and Positive Raw Stock 



*. U.S. Pot. Off. 

PH22.72-1950 

(Z22.72-1950) 



Page 2 of 2 Page* 






Appendix 



The dimensions given in this standard represent the practice of film manu- 
facturers in that the dimensions and tolerances are for film immediately after 
perforation. The punches and dies themselves are made to tolerances con- 
siderably smaller than those given, but owing to the fact that film is a plastic 
material, the dimensions of the slit and perforated film never agree exactly 
with the dimensions of the punches and dies. Shrinkage of the film, due to 
change in moisture content or loss of residual solvents, invariably results in 
a change in these dimensions during the life of the film. This change is 
generally uniform throughout the roll. 

The uniformity of perforation is one of the most important of the variables 
.affecting steadiness of projection. 

Variations in pitch from roll to roll, are of little significance compared to 
variations from one sprocket hole to the next. Actually, it is the maximum 
variation from one sprocket hole to the next within any small group that is 
important. This is one of the reasons for the method of specifying uniformity 
in dimension B. 

Thirty-two-millimeter release print stock is slit, after printing and develop- 
ing, to 16-mm width. Since a possible error is involved in this slitting, the 
width of 32-mm film is made 0.001 inch narrower than twice the width of 
standard 16-mm film. This narrowing gives a tolerance of 0.001 inch in this 
secondary slitting operation. If the error of slitting exceeds this tolerance, 
one of the 16-mm halves may exceed the width allowed for 16-mm film 
and cause interference in the gate of a projector. In addition to errors of 
centering, there are errors caused by recurring variations in width. These 
errors will cause weave on the screen even though the maximum width of 
the film may not be great enough to cause interference in the projector gate. 



244) February 1951 Journal of the SMPTE Vol. 56 



Proposed American Standard 

16-Millimeter Motion Picture 
Projection Reels 

(Second Draft) 


PH22.1 1 

(Z22.ll) 




P. 1 of 3 pp. 


("W-+-AT PERIPHERY 


ai; 


/ \ W- -AT CORE 

/ 4*' ""\\ \ ENL 
/ f \ \ FLAN 

f \\ D ~w- 
I 11 8 C x AT SPINDLE 


\RGED VIEW OF HOLE N 
SE ON LEFT IN SECTIONAL 
/IEW SHOWN ABOVE 


\ \ / / T "~< *^ T 


r!^u 


t 7 u v 

B / V 
J-^-vV 


xV 
_____y_J 


L-B J ENLARGED VIEW OF HOLE IN 
FLANGE ON RIGHT IN SECTIONAL 
VIEW SHOWN ABOVE 


TABLE 1 




Dimension Inches 


Millimeters 


+0.000 
A ' 319 -0.003 


8 - 10 1SS 


+0.000 
' 319 -0.003 


8 -' ^.M 


*i R 1 0.790 maximum 


20.06 maximum 


S 2 (including flared, 
rolled, or beveled 0.962 maximum 
edges) 


24.43 maximum 


T (adjacent to 0.027 minimum 
spindle) 0.066 maximum 

U 6.312 0.016 


0.69 minimum 
1.68 maximum 

7.92 0.41 


+0.005 
' 123 -0.000 


* ss 


W, at periphery 3 0.660 ** ^ 


16 - 76 iaii 


at core 4 0.660 0.010 


16.76 0.25 


at spindle holes 0.660 0.015 


16.76 0.38 


Flange and core 0.031 
concentricity 5 


0.79 



See Notes on p. 3. 



February 1951 Journal of the SMPTE Vol. 56 



241 



Proposed American Standard 




16-Millimeter Motion Picture 


PH22.1 1 


Projection Reels 


(Z22.ll) 


(Second Draft) 






f. 2 of 3 pp. 


TABLE 2 




Capacity Dimension Inches Millimeters 


Capacity Dimension 


Inches Millimeters 


200 Feet 6 D, nominal 5.000 127.00 


1200 Feet D, nominal 


12.250 311.15 


(61 Meters) maximum 5.031 127.79 


(366 Meters) maximum 


12.250 311.15 


minimum 5.000 127.00 


minimum 


12.125* 307.98* 


C, nominal 1.750 44.45 


C, nominal 


4.875 123.83 


maximum 2.000* 50.80* 


maximum 


4.875 123.83 


minimum 1 .750 44.45 


minimum 


4.625* 117.48* 


Lateral run- n __ n 
out, 7 maximum - 570 M5 


Lateral run- 
out, 7 maximum 


0.140 3.56 


400 Feet 6 D, nominal 7.000 177.80 


1600 Feet D, nominal 


13.750 349.25 


(122 Meters) maximum 7.031 178.59 


(488 Meters) maximum 


14.000* 355.60* 


minimum 7.000 177.80 


minimum 


13.750 349.25 


C, nominal 2.500 63.50 


C, nominal 


4.875 123.83 


maximum 2.500 63.50 


maximum 


4.875 123.83 


minimum 1.750* 44.45* 


minimum 


4.625* 117.48* 


ou^ma'xTmum a 80 2 ' 03 


Lateral run- 
out, 7 maximum 


0.160 4.06 


800 Feet D, nominal 10.500 266.70 


2000 Feet D, nominal 


15.000 381.00 


(244 Meters) maximum 10.531 267.49 


(610 Meters) maximum 


15.031 381.79 


minimum 10.500 266.70 


minimum 


15.000 381.00 


C, nominal 4.875 123.83 


C, nominal 


4.625 117.48 


maximum 4.875 123.83 


maximum 


4.875 123.83 


minimum- 4.500* 114.30* 


minimum 


4.625 117.48 


Lateral run- 
out/ maximum - 120 3 ' 05 


Lateral run- 
out, 7 maximum 


0.171 4.34 



242 



February 1951 Journal of the SMPTE Vol. 56 



Proposed American Standard 

16-Millimeter Motion Picture 
Projection Reels 



(Second Draft) 



PH22.11 
(Z22.ll) 



NOTES 

* When new reels are designed, or when new tools 
are made for present reels, the cores and flanges 
should be made to conform, as closely as practicable, 
to the nominal values in the above table. It is hoped 
that in some future revision of this standard the as- 
terisked values may be omitted. 

1 The outer surfaces of the flanges shall be flat out 
to a diameter of at least 1.250 inches. 

2 Rivets or other fastening members shall not ex- 
tend beyond the outside surfaces of the flanges more 
than 1/32 inch (0.79 millimeters) and shall not extend 
beyond the over-all thickness indicated by dimension S. 

3 Except at embossings, rolled edges, and rounded 
corners, the limits shown here shall not be exceeded 
at the periphery of the flanges, nor at any other dis- 
tance from the center of the reel. 

4 If spring fingers are used to engage the edges 
of the film, dimension W shall be measured between 
the fingers when they are pressed outward to the 
limit of their operating range. 

3 This concentricity is with respect to the center line 
of the hole for the spindles. 

* This reel should not be used as a take-up reel on 
a sound projector unless there is special provision TO 
keep the take-up tension within the desirable range 
of IVi to 5 ounces. 

7 lateral runout is the maximum excursion of any 
point on the flange from the intended plane of rota- 
tion of that point when the reel is rotated on an accu- 
rate, tightly fined shaft. 

APPENDIX 

Dimensions A and B were chosen to give 
sufficient clearance between the reels and the 
largest spindles normally used on 16-milli- 
meter projectors. While some users prefer a 
square hole in both flanges for laboratory 
work, it is recommended that such reels be ob- 
tained on special order. If both flanges have 
square holes, and if the respective sides of the 
squares are parallel, the reel will not be suit- 
able for use on some spindles. This is true if 
the spindle has a shoulder against which the 
outer flange is stopped for lateral positioning 
of the reel. But the objection does not apply if 
the two squares are oriented so that their re- 
spective sides are at an angle. 

For regular projection, however, a reel with 
a round hole in one flange is generally pre- 
ferred. With it the projectionist can tell at a 
glance whether or not the film needs rewind- 



P. 3 of 3 pp. 

ing. Furthermore, this type of reel helps the 
projectionist place the film correctly on the 
projector and thread it so that the picture is 
properly oriented with respect to rights and 
lefts. 

The nominal value for W was chosen to 
provide proper lateral clearance for the film, 
which has a maximum width of 0.630 inch. Yet 
the channel is narrow enough so that the film 
cannot wander laterally too much as it is 
coiled; if the channel is too wide, it is likely to 
cause loose winding and excessively large 
rolls. The tolerances for W vary. At the core 
they are least because it is possible to control 
the distance fairly easily in that zone. At the 
holes for the spindles they are somewhat 
larger to allow for slight buckling of the 
flanges between the core and the holes. At 
the periphery the tolerances are still greater 
because it is difficult to maintain the distance 
with such accuracy. 

Minimum and maximum values for T, the 
thickness of the flanges, were chosen to per- 
mit the use of various materials. 

The opening in the corner of the square 
hole, to which dimensions U and V apply, is 
provided for the spindles of 35-millimeter re- 
winds, which are used in some laboratories. 

D, the outside diameter of the flanges, was 
made as large as permitted by past practice 
in the design of projectors, containers for the 
reels, rewinds, and similar equipment. This 
was done so that the values of C could be 
made as great as possible. Then there is less 
variation, throughout the projection of a roll, 
in the tension to which the film is subjected by 
the take-up mechanism, especially if a con- 
stant-torque device is used. Thus it is necessary 
to keep the ratio of flange diameter to core 
diameter as small as possible, and also to 
eliminate as many small cores as possible. For 
the cores, rather widely separated limits (not 
intended to be manufacturing tolerances) are 
given in order to permit the use of current reels 
that are known to give satisfactory results. 



NOT APPROVED 



February 1951 Journal of the SMPTE Vol. 56 



243 



Proposed American Standard 

Dimensions for Projection Lamps 
Medium Prefocus Ring Double-Contact Base-Up Type 

for 16-Mm and 8-Mm Motion Picture Projectors 



PH22.84 




p. i of a PP. 



, 0.005 



ELECTRICAL CONTACTS 
SEE PAR 3 



THIS SIDE TOWARD 
CONDENSER LENS 



VENTILATING PORTS 
SEE PAR I 



0.094 0.005 R 



-BODY OF BASE SHALL 
PASS THRU A RING 
' 1.063 D 



1 
a 

r>- 
6 




6 
M 


r 


^Z 












i 


/ 


\ 


j 




i 

a 




n 


i 


u 






<r f 


1 










O 








X 

J).850^ 


-0-7.85^ 



THESE THREE POINTS 
REGISTER AGAINST 
FIXED SURFACE IN 
LAMP HOLDER 





CONDENSER SIDE 
OF LAMP 



-SOURCE CENTERED 
ON AXIS OF PRE- 
FOCUS RING WITHIN 
0.030, FRONT AND 
SIDE VIEWS 



ALL DIMENSIONS IN INCHES 



1. Scope. The purpose of this standard is to 
establish, for the type of lamp shown, the di- 
mensions essential to Intel-changeability of 
lamps in projectors. It is not intended to pre- 
scribe either operating characteristics or de- 
tails of design such as the shape of the ven- 



tilation ports or method of attachment of the 
prefocus ring to the base. 

2. Operating Position. Lamps of this type 
are intended to be burned with the axis in an 
essentially vertical position, and with the base 
at the top. 



NOT APPROVED 



244 



February 1951 Journal of the SMPTE Vol. 56 



Proposed American Standard 

Dimensions for Projection Lamps 

Medium Prefocus Ring Double-Contact Base-Up Type 

for 1 6-Mm and 8-Mm Motion Picture Projectors 



PH22.84 



3. Electrical Contacts. The drawing indi- 
cates the area which the electrical members of 
the lamp holder should contact. It is not in- 
tended to dictate the shape of the terminals on 
the lamp. With lamps of this type, the pre- 
focus ring is not an electrical contact. 

Note 1. These dimensions define the maximum ex- 
cursion of the bulb surfaces from the base axis toward 



P. 2 of 2 pp. 

the condensing lenses and the mirror at the points 
indicated when the lamp is inserted in a holder which 
rotationally positions the lamp as shown in the end 
view of the base. Condensing lenses, the mirror, and 
their mounts must therefore be so located as to insure 
adequate clearance between these parts and the bulb 
surface. 

Note 2. For medium prefocus base-down projection 
lamps, see PH22.85. 



NOT APPROVED 



February 1951 Journal of the SMPTE . VbL 56 



245 



Proposed American Standard 

Dimensions for Projection Lamps 
Medium Prefocus Base-Down Type 

for 16-Mm and 8-Mm Motion Picture Projectors 



PH22.85 




THIS SIDE OF BULB 
TOWARD CONDENSER 
LENS 



SEE J T -' BULB 
I T-12 BULB 



MOTH-MI* 



1 . Scope. The purpose of this standard is to 
establish, for the type of lamp shown, the di- 
mensions essential to interchangeability of 
lamps in projectors. It is not intended to pre- 
scribe either operating characteristics or de- 
tails of design. 

2. Operating Position. Lamps of this type 
are intended to be burned with the axis in an 
essentially vertical position, and with the base 
at the bottom. 



Fiery 


>-l. 548 MAX * 




t k*^> 


FOR T-12 BULB 




80^ 


-I.294MAX-* 




>P VIEW OF BASE 


FOR T-IO 






BU 


LB 




X"f\ 




V 


X 




i 


0.668 MAXl 








mi 


T-IOBULB! SEE 








\ '. / 


0.785 MAX f NOTE t 








L^ 


JU 


^-12 BULB ) 














- SOURCE CENTERED - 


_! 






^~~~~~' 




ON AXIS OF PRE- 


^*nn 


X 








FOCUS BASE WITHIN 


1 




< 








0.030, FRONT AND 


* H 




2 








SIDE VIEWS 






S 






g 
ni 






10.030 


r- 










THIS SURFACE REG- 






g 








ISTERS AGAINST 






*^ 
' 








FIXED SURFACE IN ^ 










, 






SOCKET 1 










\ 


J 






Vv 


J 






5= 




t 








a 

6 


6 


/ -'- e 

/ 0.< 


6D 
05 "*~ 


L -1.328 D. 
tO.005 


' Hr~i ^ 


^BODY OF BASE SHACL 


] ^L^ tl 






PASS THRU A^fNG OF 


' 98 D ALL DIMENSIONS IN INCHES 



Note 1. These dimensions define the maximum ex- 
cursion of the bulb surfaces from the base axis toward 
the condensing lenses and the mirror at the points 
indicated when the lamp is inserted in a holder which 
rotational!/ positions the lamp as shown in the end 
view of the base. Condensing lenses, the mirror, and 
their mounts must therefore be so located as to insure 
adequate clearance between these parts and the bulb 
surface. 

Not* 2. For medium prefocus ring double-contact 
base-up projection lamps, see PH22.84. 



NOT APPROVED 



246 



February 1951 Journal of the SMPTE Vol. 56 



69th Semiannual Convention 



PAPERS for presentation at the Spring 
Convention at the Statler Hotel in New 
York, April 30-May 4, are now being as- 
sembled by the 1951 Papers Committee. 
Committee appointments were completed 
in early February. The six Vice-Chair- 
men and all committee members are listed 
below. Members who wish to make a 
presentation at this convention or who 
know of developments which should be 
reported on promptly are requested to 
correspond directly with the proper Papers 
Committee Vice-Chairman. Each of 
these Vice-Chairmen has available copies 



of the present Author's Forms, "Hints to 
Authors" (which suggests appropriate ways 
of manuscript preparation) and copies of 
American Standard Z38.7. 19-1950 Dimen- 
sions of Lantern Slides. Be sure to contact 
your Papers Committee member promptly, 
for an early publication date for the Tenta- 
tive Program has been established. It is 
essential that these deadlines be main- 
tained so that members whose attendance 
at the convention depends upon the pres- 
entation of technical material in their own 
fields may be able to make their plans. 



PAPERS COMMITTEE 

Chairman, Edward S. Seeley, Altec Service, 161 Sixth Ave., New York 13 

Vice-Chairmen 

For New York: W. H. Rivers, Eastman Kodak Co., 342 Madison Ave., New York 17 

For Washington: J. E. Aiken, 116 N. Galveston St., Arlington, Va. 

For Chicago: R. T. Van Niman, 4441 Indianola Ave., Indianapolis, Ind. 

For Los Angeles: F. G. Albin, American Broadcasting Co., Station KECA-TV, 4151 

Prospect Ave., Hollywood, Calif. 

For Canada: G. G. Graham, National Film Board of Canada, John St., Ottawa, Canada 
For High-Speed Photography: J. H. Waddell, Wollensak Optical Co., 850 Hudson Ave., 

Rochester, N.Y. 



Committee Members 

A. C. Blaney, RCA Victor Div., 1560 N. 

Vine St., Hollywood 28, Calif. 
Richard Blount, General Electric Co., 

Nela Park, Cleveland, Ohio. 
R. P. Burns, Balaban & Katz, Great 

States Theaters, 177 N. State St., 

Chicago 1, 111. 
Philip Caldwell, American Broadcasting 

Co., 6285 Sunset Blvd., Hollywood, 

Calif. 

F. O. Calvin, The Calvin Co., 1105 E. 
Fifteenth St., Kansas City 6, Mo. 

Howard Chinn, Columbia Broadcasting 

System, 485 Madison Ave., New York 

22 
J. P. Corcoran, Twentieth Century-Fox 

Film Corp., 10201 W. Pico Blvd., 

Beverly Hills, Calif. 

G. R. Crane, Westrex Corp., 6601 Ro- 
maine St., Hollywood 38, Calif. 

E. W. D'Arcy, De Vry Corp., 1111 W. 
Armitage Ave., Chicago 14, 111. 



Farciot Edouart, Paramount Pictures 
Corp., 5451 Marathon St., Hollywood 
38, Calif. 

F. L. Eich, Paramount Film Laboratory, 
1546 Argyle Ave., Hollywood 28, Calif. 

Dudley Goodale, National Broadcasting 
Co., 30 Rockefeller Plaza, New York 20 

Charles Handley, National Carbon Div., 
841 E. Fourth PI., Los Angeles 13, Calif. 

R. N. Harmon, Westinghouse Radio Sta- 
tions, Inc., 1625 K St., N.W., Washing- 
ton, D.C. 

Scott Kelt, Allen B. Du Mont Labs., Inc., 
2 Main Ave., Passaic, N.J. 

C. E. Heppberger, National Carbon Div., 
230 N. Michigan Ave., Chicago 1, 111. 

J. K. Hilliard, Altec Lansing Corp., 1161 
N. Vine St., Hollywood 38, Calif. 

L. Hughes, Hughes Sound Films, 21 S. 
Downing St., Denver, Colo. 

P. A. Jacobson, University of Washington, 
Seattle, Wash. 



247 



William Kelley, Motion Picture Research 
Council, 1421 N. Western Ave., Holly- 
wood 27, Calif. 

E. P. Kennedy, Signal Corps Labs, Fort 
Monmouth, N.J. 

George Lewin, Signal Corps Photographic 
Center, 35-11 35 St., Long Island City 
1, N.Y. 

E. C. Manderfeld, Mitchell Camera Corp., 
666 W. Harvard St., Glendale 4, Calif. 

Glenn Matthews, Research Laboratory, 
Eastman Kodak Co., Rochester 10, N.Y. 

Pierre Mertz, Bell Telephone Labs., Inc., 
463 West St., New York 14 

James Middlebrooks, American Broadcast- 
ing Co., 30 Rockefeller Plaza, New York 
20 

Harry Milholland, Allen B. Du Mont 
Labs, Inc., 515 Madison Ave., New 
York 22 

W. J. Morlock, General Electric Co., 
Electronics Park, Syracuse, N.Y. 



Herbert Pangborn, Columbia Broadcasting 
System, Inc., 6121 Sunset Blvd., Holly- 
wood 28, Calif. 

Edward Schmidt, Reeves Soundcraft, 10 
E. 52 St., New York 22 

N. L. Simmons, Eastman Kodak Co., 
6706 Santa Monica Blvd., Hollywood 
38, Calif. 

S. P. Solow, Consolidated Filin Industries, 
Inc., 959 Seward St., Hollywood 38, 
Calif. 

J. G. Stott, Du-Art Film Laboratories, 
245 W. 55 St., New York 19 

W. L. Tesch, Radio Corporation of 
America, RCA Victor Div., Front and 
Cooper Sts., Camden, N.J. 

S. R. Todd, Consulting Electrical Engi- 
neer, 4711 Woodlawn Ave., Chicago, 111. 

M. G. Townsley, Bell & Howell, 7100 
McCormick Rd., Chicago 45, 111. 



Board of Governors Meeting 



PROGRESS was the keynote of the Annual 
Meeting of the Society's Board of Gover- 
nors held in New York on January 24. 
The Chairman was Peter Mole, Society 
President, who took office the first of this 
year. 

The Board first reviewed the Society's 
business, .technical and publications activ- 
ities for the year 1950, hearing reports by 
several officers. Since Ralph B. Austrian, 
Financial Vice-President was unable to at- 
tend the meeting, Frank E. Cahill, Treas- 
urer, presented the report of business activ- 
ities and a summary of the financial 
status of the Society as of December 31, 
1950. His analysis of operations showed 
that 1950 was the Society's busiest year but 
that although the Society's income was up, 
administrative expenses were down, being 
actually slightly lower than the previous 
year. 

ENGINEERING 

Technical activities were reported by 
Fred T. Bowditch, Engineering Vice- 
President. Among the items which he re- 
viewed was the much discussed question of 
whether the Society should or should not 
formally request that the American Stand- 



ards Association send out a call for a meet- 
ing of technical .committee TC-36, Cine- 
matography, of the International Stand- 
ardization Organization (ISO). Such a 
meeting had been proposed for this com- 
ing summer in Switzerland and it was felt 
that the Society's position of responsibility 
in this connection made a prompt decision 
mandatory. After soliciting the opinions 
of several Society members who had been 
seriously interested in standards on an 
international basis for many years, the 
Board of Governors concluded that calling 
such a meeting at this time would be quite 
inefficient and would represent consider-, 
able waste of time and money because no 
specific agenda had been developed. The 
Board did recommend, however, that 
vigorous attention be given to the inter- 
national standards picture by the ASA 
Sectional Committee on Motion Pictures 
and by our own standards and technical 
committees. A responsible agenda could 
doubtless be developed in time for the ISO 
meetings already scheduled for 1952 in the 
United States. 

The Board voted sponsor approval of 
three proposed American Standards which 
subsequently require the approval of 



248 



ASA's Photographic Standards (Corre- 
lating) Committee and also of the ASA 
Board of Review. They cover Dimensions 
for 32/ 35-mm Negative StocK;%ociis Base 
Point -for 16- arid^-mm Cameras; : afe 
Threads and Flange Distances fo j: l 6- and 
8-mm Camera Lenses. 

:,.. . , . . -:-.,, (i . 

PUBLISHING 

Since neither 01yde E. Keith, former 
Editorial Vice-President, nor John G. 
Frayne, Editorial Vice-Presicleiit for 1951- 
52, was able to attend the meeting, their 
written reports were offered by the Secre- 
tary. Mr. Frayne announced reappoint- 
ment of Arthur C. >6wnes as Chairman of 
the Board of Editors, and observed that 
the Society was particularly fortunate in 
having had such a capable engineer take a 
great personal interest in the Society's 
JOURNAL. He announced also that Ed- 
ward S. Seeley had been appointed Na- 
tional Chairman of the Papers Committee 
and that he would be assisted during 1951 
by five regional Vice-Chairmen and one 
Vice-Chairman representing a specialized 
industry group (see Convention story 
above). 

Charles W. Handley had been reap- 
pointed Chairman of the Progress Com- 
mittee and Edmund A. Bertram had ac- 
cepted the Chairmanship of the Historical 
and Museum Committee. Mr. KeitlT ob- 
served that the budget limitations during 
the last three months of 1950 had lorced a 
slight reduction in the amount of material' 
published in the J6URNAL. Changes 
which he, Mr. Frayne and Victor 'Allen 
had agreed on for JOURNAL format would 
improve appearance of the JOURNAL and it 
was noted that some budget relief might be 
achieved. Two other changes were the 
addition of abstracts of technical articles 
published elsewhere but not generally 
available to members, and the preparation 
of semitechnical reports of local Section 
meetings which were of serious interest to 
Society members but for which no formal 
manuscript had been prepared by the 
speaker. It was suggested that members 
in the three local Sections might be per- 
suaded to prepar^ guch reports regularly. 

RECORDING DISCUSSIONS 

Over the ye&ts'j many attempts were 
made to record the discussions which fol- 



low presentation of technical papers at 
Society Conventions. Stenotype operators 
had 'been employed, as well as recording 
rftefiods whicli used embossed tape, discs 
and quarter-inch magnetic tape. During 
the convention in Lake Placid last year, a 
professionaFniodel of magnetic recording 
machine was loaned to the Society and 
proye<Ho yfelcphe best-results achieved so 
fslr; Mr. Keith suggested that a similar 
method, somewhat simplified, might well 
be adopted as a permanent system for re- 
cording, since the yield was higher and the 
cost somewhat lower than when using a 
stenotype operator. The Board accepted 
this suggestion and authorized Mr. Keith 
to proceed with the design and acquisition 
of suitable equipment which would be 
owned by the Society and maintained '^ 
the Headquarters Staff. 

Another project that had been under 
Mr. Keith's supervision was the design of a 
more symbolic emblem for the Society to 
replace the one currently used on Society 
letterhead. Mr. Keith resigned as Chair- 
man of that committee and Mr. Mole was 
asked to discuss the further activity alongs 
those lines with Lorin D. Grignon. 

CONVENTIONS 

In reviewing the 67th and 68th Conven- 
tions held in Chicago and Lake Placid dur- 
ing 1950, Mr. Kunzmann reported with 
some enthusiasm that income from regis- 
trations had offset convention expenses 
and he was able to turn in a black figure 
at the end of the year. He also reported on 
arrangements for the 69th Convention, in 
New York, as well as the 70th Convention, 
scheduled for the Hollywood-Roosevelt 
Hotel, October 15-19, 1951. 

PLANNING FOR 1951 

With reports of Society activity for 1950, 
completed, the Board considered the pro-# 
posed budget for 1951 and endorsed a 
program of expansion, which included cur- 
rent growth in every phase of the Society's 
work. Space limitations at Society Head- 
quarters have hampered engineering and 
publications activities to an extent that 
began to assume serious proportions. In 
recognizing this problem, the Board 
authorized the Executive Secretary to ac- 
quire larger quarters and to employ two 
additional staff members. One will pro- 



249 



vide additional stenographic assistance and 
the other will be a young engineer, assigned 
almost entirely to the Society's Test Film 
Program. His interests would be along the 
lines of quality control, test film production 
and the more accurate specification of 
particular films, either now being made or 
proposed for the future. 

Expansion of publications activity by 
72% was authorized and it was pointed 
out that the change in JOURNAL format 
would allow the editor to purchase the 
same amount of printed information for 
each dollar during 1951 as he did during 
the early months of 1950, even though 
general publications costs had increased 
considerably. Demand for additional 
technical work by Headquarters and by 
many of the committees seemed completely 



justified and an increase of the services 
thus provided by as much as 100% was 
authorized. 

Membership procedures have been sim- 
plified so that a fairly large increase in the 
rolls could be handled by Headquarters 
with very little additional effort. This 
increased efficiency makes it practical for 
the Society to invite membership from 
many potential applicants who have not 
been aware of the services available. 
Headquarters has been requested to apply 
additional effort along these lines, with the 
hope that every potential candidate for 
membership will learn of the Society and 
its JOURNAL and as a consequence will be 
able to judge whether the benefits warrant 
his joining. 



1951 Nominations 



'VOTING' members of the Society, that is 
all those in the Active, Fellow and Honor- 
ary grades, are invited by the Chairman of 
the Nominating Committee to suggest 
candidates for the seven Board of Gover- 
nors' vacancies which will occur at the end 
of 1951. Since the Nominating Committee 
for this year will soon begin its formal de- 
liberations, names of potential nominees 
should be placed in the hands of the Chair- 
man, Earl I. Sponable, c/o Movietonews 
Inc., 460 W. 54 St., New York 19, without 
delay. 



Vacancies will be for the offices held now 
by the following members of the Board of 
Governors. The only limitations on sug- 
gested candidates for these vacancies are 
that they be Voting members of the 
Society. 

Financial Vice- President, Ralph B. Aus- 
trian 

Treasurer, Frank E. Cahill, Jr. 

Governors, Lorin D. Grignon, Paul J. 
Larsen, William H. Rivers, Edward S. 
Seeley and R. T. Van Niman 



Engineering Activities 



Magnetic Recording 

In the September, 1950, JOURNAL, it was 
reported that the proposals for Magnetic 
Sound Track on Film, originating in Glenn 
Dimmick's Subcommittee, were meeting 
obstacles in the parent Sound Committee. 
Those obstacles have been overcome and 
the proposals are now before the members 
of the Standards Committee, who are now 
balloting on their recommendation for 
preliminary publication for trial and com- 
ment. 



Laboratory Practices 

The Laboratory Practice Committee, 
under the Chairmanship of John Stott, 
met in January and pushed ahead on its 
ambitious program. 

The work on 35-mm negative notching 
is reaching a conclusion and the com- 
mittee will soon be canvassed on approval 
of the draft specification for size and 
location of notches. 

16-Mm negative notching has presented 
a more thorny problem which will require 



250 



industry-wide assistance for solution. To 
this end, an interim committee report is 
being prepared for publication, with the 
expectation that sufficient comments will 
thus be elicited to enable the writing of 
an adequate draft specification. 

A report of the Screen Brightness Survey 
of the 16-mm review rooms of the film 
processing laboratories was submitted. 
On the basis of the data accumulated at 
both East and West Coast laboratories, it 



was recommended that a 16-mm screen 
brightness standard be drafted, and con- 
crete proposals to that effect are in the 
making. 

The committee has been planning for 
some time to provide abstracts of chemical 
engineering material for a regular page in 
the JOURNAL. Proposals to achieve this 
were discussed and responsibilities fixed. 
We can, therefore, expect this valuable 
service to be initiated shortly. 



Obituaries 



Joseph Mina Bing, who was an influential 
force in amateur photography and in the 
photographic industry, died at his home in 
New York on December 9, 1950. He was 
72 years old. He was born in Vienna and 
was graduated from the University of 
Vienna with the degree of Doctor of Engi- 
neering. He was engaged in consulting 
railroad work in Austria, South and Cen- 
tral American Countries and in the build- 
ing of the Hell Gate Bridge. 

In 1925 he became the first importer of 
photographic exposure meters, in which 
field he was an expert and designer. He 
later became one of the largest importers 
of cameras and other equipment. During 
World War II his manufacturing organi- 
zation received two Army-Navy awards for 
its excellent work in producing Navy test- 
ing equipment and design of the under- 
water camera. Mr. Bing was an Honorary 
Fellow of both the Royal Photographic 
Society and the Photographic Society of 
America. He had been an Active Member 
of this Society for 22 years. 

Lewis M. Townsend died on October 16, 
1950. He had long been an active member 
in the Society. In 1925 he coauthored 
with L. A. Jones a paper "The Use of 
Color for the Embellishment of the Mo- 
tion Picture Program," which was pub- 
lished in the TRANSACTIONS of the Society. 
He was also coauthor of many other papers. 
For several years he was Chief Projection 
Engineer of the Eastman Theatre and 
the School of Music at the University of 
Rochester. He was Technical Adviser on 
Sound Equipment for Paramount Publix 
from 1929 to 1932 and from 1932 until his 



death he was Chief Projectionist and Head 
of Equipment Maintenance for Schine 
Theaters, Inc. 

Jack E. Beach was killed in a plane that 
crashed on Mt. Moran, Wyoming, No- 
vember 21. He was 23 years old. He had 
worked as an assistant cameraman for 
Coronet Studios and as cameraman for C. 
O. Baptista Films, before being appointed 
to the staff of the New Tribes Mission as 
Production Manager in charge of their 
film work, which was to have been an ex- 
position of missionary activities in the 
South. It was a mission-owned plane, des- 
tined for Florida, Bolivia and Brazil, on 
which he was killed. 



Joseph W. Fleming, Manager of the Tech- 
nical Information Center for Philips Labo- 
ratories, Irvington-on-Hudson, N.Y., was 
killed in an automobile accident on Febru- 
ary 12, near his home in Edgewater, N.J. 
He had been an IATSE Member and had 
his own radio and sound business from 1929 
to 1942 when he became associated with 
National Broadcasting Co. as Sound and 
Maintenance Engineer. Well known in 
radio and television, Mr. Fleming had 
served overseas in World War II as tech- 
nical adviser to the U.S. Air Force in 
Europe and the Royal Air Force. At that 
time, he was also attached to the British 
Ministry of Aircraft Production. He had 
been an Active Member of this Society 
since 1947. He was also a member of the 
Audio Engineering Society, Institute of 
Radio Engineers and Photographic Society 
of America. 



251 



New Members 



The following members have been added to the Society's rolls^ince those published last month. 
The designations of grades are the same as those used in the 1950 MEMBERSHIP DIRECTORY. 



Honorary (H) 



Fellow (F) 



Active (M) 



Associate (A) 



Student (S) 



Babb, Harry L., Salesman, Eastman Kodak 
Co. Matt: 710 Crenshaw Blvd., Los 
Angeles 5, Calif. (A) 

Barton, Cecil W., Electrical Technician, 
Universal Pictures Corp. Mail: 13924 
Weddington St., Van Nuys, Calif. (M) 

Bernd, Lester E., 16-Mm Motion Picture 
Producer, Delaware Steeplechase & Raise 
Association. Mail: 11 Comeau St., 
Wellesley Hills 82, Mass. (M) 

Bieling, Robert O., Head, Film Quality 
Control, Bell & Howell Co. Mail: 96 
Commonwealth Rd., Rochester 18, N.Y. 
(M) 

Blaskiewicz, John V., New Institute for 
Film and Television. Mail: 625 Hinsdale 
St., Brooklyn 7, N.Y. (S) 

Braaten, Norman J., Audio-Visual Elec- 
tronics Technician, Minneapolis Board of 
Education. Mail: 5152 29 Ave., S., 
Minneapolis 17, Minn. (A) 

Brueckner, Hubert U., Superintendent, 
Optical Shop, Revere Camera Co. Mail: 
1117 S. East Ave., Oak Park, 111. (M) 

Cambi, Enzo, Consulting Engineer, Cine- 
citta Studios; Lecturer, National Re- 
search Council (Italy) and Leghorn Naval 
Academy. Mail: Via Giovanni Antonelli 
3, Rome, Italy. (A) 

Cheda, Paul M., Artist. Mail: 90-47 53 
Ave., Elmhurst, L.I., N.Y. (A) 

Chen, Aegem, University of Southern Cali- 
fornia. Mail: 2610^ N. Broadway, Los 
Angeles 31, Calif. (S) 

Cole, Henry James, Photographer, National 
Institutes of Health, U.S. Public Health 
Service. Mail: 212 Piping Rock Dr., 
R.F.D. #2, Silver Spring, Md. (M) 

Day, James A., Projection Supervisor, 
WXYZ-TV, Inc. Mail: 12768 Elgin Ave., 
Huntington Woods, Mich. (A) 

Duncan, Cyril J., Director, Dept. of Photog- 
raphy, University of Durham, King's 
College, Newcastle Upon Tyne, 1, Eng- 
land. (M) 

Franzen, Russell G., Industrial Photog- 
rapher, American Can Co. Mail: 1412 
S. Fourth Ave., Maywood, 111. (M) 

Fussel, Alex, Theater Projectionist, Valus- 
kis Theatres. Mail: 11914 Cheshire St., 
Norwalk, Calif. (A) 

Gibbons, Thomas J., Jr., Sales Engineer, 
Minnesota Mining & Manufacturing Co. 



Mail: 909 Kingsley Dr., Arcadia, Calif. 
(M) 

Gordon, James B., Director of Photography 
and Head of Optical Printing Dept., 20th 
Century-Fox Film Corp. Mail: 2225 
Linnington Ave., Los Angeles 64, Calif. 
(M) 

.Graff, Lee, Chief Engineer, Brenkert Light 
, -Projection Co. Mail: 4510 Lawre Rd., 
Centerline, Mich. (M) 

Heirabach, Newton, Chemist, Assistant Film 
Plant Manager, Bell & Howell Co. Mail: 
145 Commonwealth Rd., Rochester, N.Y. 
(M) 

Hoff, J. Robert, Sales Manager, The Ballan- 
tyne Co. Mail: 1707 Davenport St., 
Omaha, Nebr. (M) ,, ; , 

Howard, Bruce, Audio Facilities Engineer- 
ing & Recording Supervisor, 'Radio 
Station WBAP, (AM-FM-TV). Mail: 
2754-B Primrose, Fort Worth, Tex. (M) 

Huntsman, Harold F., Television Engineer- 
ing Field Supervisor, KECA-TV. Mail: 
409 Irving Ave., Glendale 1, Calif. (M) 

Hyll, Richard, Laboratory Technician, Film 
Associates. Mail: 325 N. Main St., 
Bowling Green, Ohio. (A) 

Iwerks, Donald, Assembly Dept., Photo- 
graphic Products. Mail: 15153^ Dickens 
St., Sherman Oaks, Calif. (A) 

Jayne, Stuart T., Assistant Plant Engineer, 
Pathe Industries, Inc. Mail: 1259 N. 
Mansfield, Hollywood 38, Calif. (A) 

Johnson, Elisha H., School of Public Rela- 
tions, Boston University. Mail: 177 
Academy St., Jersey City 6, N.J. (S) 

Johnson, Stanley L., University of Southern 
California. Mail: 1041 Browning Blvd., 
Los Angeles 37, Calif, (S) 

Kotis, Arnold F. T., Free-Lance Camerman. 
Mail: 39-37 49 St., Long Island City 4, 
N.Y. (A) 

Krause, Edward B., Manufacturer of Film 
Processing Equipment. Mail: 40 Birch 
PL, Stratford, Conn. (M) 

Lawrence, Robert L., Eastern Manager, 
Jerry Fairbanks, Inc. Mail: 235 E. 73 
St., New York 21, N.Y. (M) 

Mabuchi, Osamu, Electronic Engineer, 

, Chemist, Matsushita, -^lectric Industrial 

!,' Co., lytd. Mail: 3S Shimam&chi, Nishi- 

Tcujo Shimosyo-ku, Kyoto - City, Japan. 

(A) 



252 



Mil witt, William, Engineer, RCA Labora- 
tories. Mail: 620 S. Catalina St., Los 
Angeles 5, Calif. (A) 

Mueller, Gustave M., Supervisor, Machine 
Shop, Pathe Laboratories, Inc. Mail: 
2913 Marsh St., Los Angeles 39, Calif. (A) 

Nebbia, Michael, Free-Lance Cinematog- 
rapher. Mail: 831 Lexington Ave., New 

- York 21, N.Y. (A) 

Oerfel, John T., Motion Picture Laboratory 
Technician, George W. Colburn Labora- 
tory. Mail: 701 Willow St., Chicago 14, 
111. (A) 

Parker, Jackson G., Motion Picture Dept., 
University of California at Los Angeles. 
Mail: 476 Midvale, Los Angeles 24, Calif. 
(S) 

Plass, Joseph P., Photographer, National 
Institutes of Health, U.S. Public Health 
Service. Mail: 4701 MacArthur Blvd., 
N.W., Washington, D.C. (M) 

Racies, Larry, Camerman, Newsreel Serv- 
ice. Mail: 140 E. 46 St., New York 17, 
N.Y. (M) 

Richards, Balfour A., District Engineer, 
Palace Amusement;, Co., (1921) Ltd., 
P.O. Box 211, Kingston, Jamaica, British 
West Indies. (A) 

Riddle, William R., Television Film Editor, 
WOR-TV. Mail: 91 Central Park West, 
New York 23, N.Y. (A) 

Roberts, Paul M., New York University. 
Mail: 29 Wadsworth Ave., New York 33, 
N.Y. (S) ,- 

Romans, William E., American Television 
Inst. Mail: 4030 N. Sheridan Rd., 
Chicago 13, 111. (S) 

Schwarz, George, Plant Manager, Rochester 
Film Div., Bell & Howell Co. Mail: 
90 Browncroft Blvd., Rochester 9, N.Y. 
(M) 

Scott, David C., Producer. Mail: 636 Las 
Casas Ave., Pacific Palisades, Calif. (A) 

Seaman, Gerald, Hollywood University. 
Mail: 1411^ N. Alexandria, Hollywood 
27, Calif. (S) 

Shea, Robert P., Mechanical Engineer, 
Producers Service Co. Mail: 5447 Rad- 
ford Ave., North Hollywood, Calif. (M). 

Sheldon, John L., Research Physicist, 
Corning Glass Works. Mail: 112 E. 
Third, Corning, N.Y. (M) 
Sinnett, Robert J., Chief Engineer, WHBFi 
AM/FM/TV, Rock Island Broadcasting 
Co. Mail: 3201 Twenty-Fifth St., Rock 
Island, 111. (M) 

Spielvogel, Bert, Motion Picture Camera- 
man, City of New York, WNYC-TV. 
Mail: 30 W. 105 St., New York 25, N.Y. 
(A) 



Swedlund, Lloyd E., Electrical Engineer, 
Radio Corporation of America, RCA 
Victor Division, Lancaster, Pa. (A) 

Tchakmaknian, Krikor, Sound Engineer, 
Nahas Studios, Pyramids Rd., Cairo, 
Egypt. (M) 

Thomas, Philip F., Test Engineer, Western 
Electric Co. Mail: 5400 Columbus Ave., 
Van Nuys, Calif. (A) 

Valentine, Christian, Jr., Art Director, 
Biow Co. Mail: 36-40 Bowne St., 
Flushing, N.Y. (A) 

Van der Wyk, Jack, University of Southern 
California. Mail: 2739 Morningside St., 
Pasadena 8, Calif. (S) 

Volmar, Victor, Publicity Director & Super- 
visor of Foreign Versions, Monogram 
International Corp. .Mail: 55 Payson 
Ave., New York 34, N.Y. (A) 

Wade, Roger W., Photographer, Motion 
Picture Producer, Roger Wade Produc- 
tions. Mail: 77-17 247 St., Bellerose, 
N.Y. (M) 

Weiss, Harry Allan, Sound Technician, 
Ryder 16-Mm Services. Mail: 6126 
Orange St., Los Angeles 48, Calif. (A) 

White, Lyman R., University of Southern 
California. Mail: 1066 W. 34 St., Los 
Angeles 7, Calif. (S) 

CHANGES IN GRADE 

Calhoun, John M., Chemist, Dept. of Manu- 
facturing Experiments, Kodak Park, 
Eastman Kodak Co., Rochester, N.Y. 
(M) to (A) 

Jamgochian, Matthew, Teacher, Los Angeles 
City Schools. Mail: 318 Road's End, 
Glendale 5, Calif. (S) to (A) 

Law, Edgar, Chief Re-Recording Engineer, 
British Lion Studio Co., London Films 
Studio, Shepperton, Middlesex, England. 
Mail: 19 Delta Rd., Worcester Park, 
Surrey, England. (M) to (A) 

Shapiro, Melvin, Editor, Projectionist, Pro- 
duction Assistant, Ryder 16-Mm Services. 
Mail: 823 N. Genesee, Los Angeles 46, 
Calif. (S) to (A) 

Somes, George W., Sound Recording Tech- 
nician, U.S. Navy Electronics Laboratory. 
Mail: 1247 Savoy St., San Diego 7, Calif. 
(A) to (M) 

Willis, John B., Free-Lance Cameraman. 
Mail: Box 1567, Grand Coulee, Wash. 
(S) to (A) 

DECEASED 
Brake, Alan R., 143 S. Robertson Blvd., Los 

Angeles 3, Calif. (A) 
Paolillo, Vincent, Field Engineer, Capitol 

Motion Picture Supply Corp., 630 Ninth 

Ave., New York (A) 



253 



BOOK REVIEWS 



The Great Audience 

By Gilbert Seldes. Published (1950) by 
Viking Press, 18 E. 48 St., New York 17. 
229 pp. 5Y 2 X 8^ in. Price $3.75. 

Twenty-five years ago Gilbert Seldes in 
The Seven Lively Arts presented the then 
daring proposition that the popular arts 
movies, radio, comic strips, vaudeville, 
etc. should be assessed by the same crit- 
ical standards which apply to the fine 
arts. He contended that the influence of 
such widely popular artists as Chaplin, 
Gershwin or Herriman merited serious 
esthetic consideration. Now, in his latest 
work, The Great Audience, he has reas- 
sessed the position of the mass arts and re- 
situated them in a larger social frame of 
reference, feeling that, while their relation 
to the fine arts is now secure, the dominant 
mass entertainments radio, movies and 
television have taken on an additional 
significance as media of mass communica- 
tion. 

To his task Mr. Seldes brings an unusual 
combination of qualifications a lively and 
sympathetic affection for the popular arts, 
bolstered by wide practical experience in 
television, radio and the movies, to- 
gether with an incisive critical temper 
unencumbered by political or intellectual 
prejudices; Out of the tremendous up- 
heaval and chaos resulting from the reloca- 
tions now taking place in the entertain- 
ment field he has not only grasped and 
clearly analyzed the significant organiza- 
tional and distribution problems, but has 
also offered reasonable proposals for re- 
organization of our existing political and 
economic framework. 

He recognizes that the popular arts have 
certain characteristics which set them off 
from the fine arts, notably the fact that 
they are not made for the ages, but created 
to be quickly enjoyed and forgotten. 

In his analyses of present-day conditions 
in the movie industry, Seldes proposes re- 
organization of the industry to meet the 
threats of television and diminishing box 
office receipts. He upsets many widely 
touted beliefs of the movie industry, such 
as "Nearly everyone goes to the movies," 
"the movie industry is the fourth largest 
in the U.S.," and that most of the profits 



of the industry come from production and 
releasing of movies. He points out that 
the manufacture of motion pictures actu- 
ally ranks about forty-sixth, that the great 
bulk of movie profits comes from distri- 
bution and exhibition rather than from 
manufacturing, and that public opinion 
polls made for the industry have indicated 
many startling facts as to who actually 
goes to the movies today. The great 
majority of those who go to the movies 
today are under twenty years of age. 
After twenty-five, people gradually stop 
going to movies geared to adolescent tastes. 
Between thirty and forty, more than half 
the population of the U.S. does not go to 
the movies more than once a month, while 
after fifty half the population never goes 
at all. Thus the claimed eighty million 
paid attendances a week actually represent 
between thirteen and fifteen million indi- 
viduals. 

Movie manufacturers, faced by the loss 
of overseas markets, the inroads on a joint 
audience by television, and the enforced 
separation of manufacturing and exhibi- 
tion facilities under antitrust regulations, 
find themselves in a precarious position. 
Part of the trouble lies in the standardized 
movie product itself, which Seldes claims 
has degenerated from the telling of a story 
to being the embodiment of a popular my- 
thology gauged for the taste of the peren- 
nially adolescent movie audience, itself 
predicated on a fixed rate of birth and age 
turnover. Seldes proposes that the movies 
try to recapture their lost audiences by 
production of more varied and mature 
films and by increasing secondary chan- 
nels of distribution for them along the 
lines of the "art" (or "sure seater") thea- 
ter and the now vanished newsreel theater 
circuits. The financial success of "Ham- 
let" has proved that even a serious "class" 
film can, with proper exploitation and dis- 
tribution, provide adult movie fare and at 
the same time make money. 

In the competition of movies and tele- 
vision for the same mass audience, Seldes 
presumes three courses of action: (1) a 
merger of interests whereby the movie pro- 
ducers could make special films for tele- 
vision and movies for their own theaters, 
and use the latter for showing both films 



254 



and television features, (2) compromise 
Hollywood may become a special manu- 
facturing unit for television, at the same 
time making films for more mature and 
specialized movie audiences, (3) active 
competition the movie industry might 
concentrate on action films, westerns and 
Technicolor musical extravaganzas where 
television cannot successfully compete. 

Seldes believes television can be most 
effective in straight dramatic productions, 
where the artificial is immediately obvious 
and out of place, as well as in its presenta- 
tion of sports and vaudeville. By combin- 
ing live news events and newsreel film a 
unique documentary form might be de- 
veloped. In an effective reorientation of 
the popular arts, television may capture 
the mass audience by presenting sports and 
vaudeville, radio may survive by present- 
ing documentary, cultural and musical 
features, and movies by concentrating on 
the production of fiction and extrava- 
ganzas. 

Provocative, thoughtful and well docu- 
mented, The Great Audience is surely one 
of the most intelligent and searching exam- 
inations of the popular arts to appear in 
many years, significant for the industrial 
specialist and the general reader alike. 
To Seldes, the popular arts are of enormous 
significance in the culture of a democracy, 
and their development and control should 
be the concern of every citizen, since we are 
all in some degree affected by them whether 
we are aware of it or not. THOMAS BARRY 
HUNT, 752 Greenwich St., New York 14. 



Movies for TV 

By John H. Battison. Published (1950) 
by Macmillan, 60 Fifth Ave., New York. 
376 pp. + numerous illus. 5% X 8^ in. 
Price $4.25. 

The recent growth of television has pro- 
vided employment for many new people 
who find themselves in an unfamiliar world. 
Also, many old hands in advertising and 
the theater feel the lack of basic informa- 
tion on a new art and science. It is to 
these people that this book is directed, 

To quote the jacket, "This book . . . 
provides information both on technical 
equipment and on program planning, 
needed to insure the best results from 



movies on television, including a great deal 
of experienced advice on technical and 
artistic details which may cause trouble." 

On page 128 the author says, "But the 
reader of this book will not normally be 
expected to have much to do with the tech- 
nical side." And on page 246, "This book 
is not intended to produce engineers, pro- 
ducers, or even technicians, but after read- 
ing and studying it the reader should be 
well prepared for any job in the film de- 
partment of a television station that does 
not require specialized technical knowl- 
edge. It should be equally helpful to any- 
one else who is concerned with films for 
television." 

Obviously any single book which treats 
such a wide variety of subjects must touch 
upon each rather lightly. But for the 
person who wishes to gain a quick broad 
view of these subjects, Mr. Battison pre- 
sents it with an easily read seminarrative 
style that is clear and pleasing. C. L. 
TOWNSEND, National Broadcasting Co., 
30 Rockefeller Plaza, New York. 



Preparation and Use of Audio- 
Visual Aids 

By Kenneth B. Haas and Harry Q. Packer. 
Published (1950) by Prentice-Hall, Inc., 
70 Fifth Ave., New York 11. i-xii + 327 
pp. including 36 pp. appendix and 7 pp. 
index. Profusely illus. 6 X 9 in. Price 
$4.65. 

This is the second edition of a well- 
known textbook. The first edition was 
"aimed at industrial and store personnel 
trainers." In this revision the authors 
attempted to broaden its appeal and use- 
fulness, but much of the original plans re- 
mains. 

The authors have compressed into a 
relatively few pages an enormous amount 
of information about every type of audio- 
visual aid. The book contains many lists 
of criteria for materials and rules on utili- 
zation of them. Although the authors keep 
the viewpoint intensely topical and devote 
very little space to theoretical considera- 
tions such as the psychology behind the 
use of audio-visual materials, the book is 
so complete that it constitutes a reference 
work in the field. All of the well-known 
instructional aids and some of the less 
well known are included. Two of the best 



255 



chapters are on using the blackboard and 
on setting up and operating an audio- 
visual laboratory. The chapter on the 
laboratory will be especially useful to 
schools of education. The authors have 
made good use of line drawings to enliven 
the text. Since the illustrations are so 
good it is surprising to find no examples of 
charts, such as pie charts and bar graphs. 
The book is so full of ideas and hundreds 
of practical suggestions that the compres- 
sion necessary to get these into a small 



space has resulted in some subjects being 
slighted. -'Tt i* impossible, obviously^ to 
explain photography in two or three pages 
or outline objective research methods in 
one page. This book, therefore, is not a 
thorough discussion of any one phase of 
audio-visual education but is an over- 
view of the whole subject. There is an 
appendix on sources of materials and 
equipment. PAUL R. WENDT, College of 
Education, University of Minnesota, Min- 
neapolis 14. 



Current Literature 



THE EDITORS present for convenient reference a list of articles dealing with subjects 
cognate to motion picture engineering published in a number of selected journals. 
Photostatic or microfilm copies of articles in magazines that are available may be 
obtained from The Library of Congress, Washington, D.C., or from the New York 
Public Library, New York, N.Y., at prevailing rates. 



American Cinematographer 

vol. 31, no. 10, Oct. 1950 
Choosing a 16Mm Camera for Professional 

Work (p. 342) L. ALLEN 
New Technicolor System Announced (p. 

354) 

vol. 31, no. 11, Nov. 1950 
Economy Prime Factor in Producing 

Films for TV (p. 377) H. A. LIGHTMAN 
Advantages of Variable Shutters in 16Mm 

Cine Photography (p. 386) J. FORBES 

vol. 31, no. 12, Dec. 1950 

New Technicolor System Tested by 

Directors of Photography (p. 414) L. 

ALLEN 

Surgical Cinematography (p. 417) F. C. 

ELLS 

New Camera and Tripod Carrier De- 
veloped at MGM (p. 418) F. FOSTER 

British Kinematography 

vol. 16, no. 4, Apr. 1950 

Technical Requirements of a Mobile Studio 
Unit for Feature Films (p. 109) B. 
HONRI 

Modern Kinema Equipment: III., Ac- 
cessory Equipment and Film Mutila- 
tion (p. 122) R. A. RIGBY 

Improvements in Large-Screen Television 
(p. 126) T. M. C. LANCE 

vol. 16, no. 5, May 1950 
Maintenance of 16Mm Print Quality 

I. The Renter's Problems (p. 152) E. 
F. BRADLEY 

II. Problems in the Field (p. 154) M. 
RAYMOND, JR. 



vol. 16, no. 6, June 1950 
High-Diffusion Screens for Process Pro- 
jection (p. 189) H. McG. Ross 

vol. 17, no. 2, Aug. 1950 
Science and the Motion Picture (p. 42) R. 

WATSON-WATT 
The Evolution of the Newsreel 

I. Introduction (p. 47) H. THOMAS 

II. The Early Days of Newsreels (p. 
47) K. GORDON 

III. The Development of the Sound 
Newsreel (p. 50) W. S. BLAND 

IV. The Future of the Newsreel (p. 53) 
H. THOMAS 

History and Development of the Colour 
Film (p. 57) R. H. CRICKS 

vol. 17, no. 3, Sept. 1950 
Electrical Devices as Applied to Special 
Effects 

I. Problems of Remote Control (p. 84) 
J.Gow 

II. Miscellaneous Equipment (p. 85) 
F. GEORGE 

Electronics 

vol. 23, no. 8, Aug. 1950 
Improved Deflection and Focus (p. 94) 
C. V. BOCCIARELLI 

vol. 23, no. 12, Dec. 1950 
Color Fundamentals for TV Engineers 
(p. 88) D. G. FINK 

vol. 24, no. 1, Jan. 1951 
Color Fundamentals for TV Engineers, 
Pt. II (p. 78) D. G. FINK 



256 



Illuminating Engineering 

vol. 45, no. 10, Oct. 1950 
Television Studio Illumination (p. 606) H. 
M. GURIN and R. L. ZAHOUB 

International Photographer 

vol. 22, no. 11, Nov. 1950 
Color of Illumination (p. 5) D. NORWOOD 
More About "Inspacian" (p. 8) I. M. 
TERWILLIGER 

vol. 22, no. 12, Dec. 1950 
TV Newsreel Production Technique (p. 5) 

J. SANDSTONE 
Smallest TV Camera (p. 8) A. L. MARBLE 

International Projectionist 

vol. 25, no. 7, July 1950 
Notes on Modern Projector Design, Pt. 

Ill (p. 5) R. A. MITCHELL 
The Ventarc H. I. Carbon 'Blown' Arc: A 

New Concept (p. 13) E. GRETENER 

vol. 25, no. 10, Oct. 1950 
L-l Arcs: Horse and Buggy Projection 

(p. 5) 
Process Projection of Film for TV (p. 8) 

R. A. LYNN and E. P. BERTERO 
Projection Shutters: A Symposium (p. 11) 

R. H. CRICKS 
The Projection of Safety Film (p. 21) R. 

A. MITCHELL 

vol. 25, no. 11, Nov. 1950 
Cinerama: Super-Movies of the Future 
(p. 10) 



Motion Picture Herald 

(Better Theatres), vol. 182, Jan. 6, 1951 
"Videofilm" Theatre TV System Using 

16Mm Stock in Production (p. 37) 
All-Plastic Screen in Radio City Music 

Hall (p. 38) 

Radio & Television News 

vol. 45, no. 1, Jan. 1951 
Servicing the 16Mm Sound Projector (p. 
36) D. D. EMERSON 

Tele-Tech 

vol. 9, no. 10, Oct. 1950 
The FCC Color TV Decision (p. 26) 

vol. 9, no. 11, Nov. 1950 

Video Recordings Improved by the Use of 

Continuously Moving Film (p. 32) W. 

D. KEMP 

vol. 9, no. 12, Dec. 1959 
Approaches to Color TV (p. 44) J. H. 

BATTISON 
High Definition Monochrome TV (p. 52) 

F. LOOMIS 
A New Video Distribution System (p. 28) 

E. D. HlLBURN 

Tele-Vision Engineering 

vol. 1, no. 10, Oct. 1950 
Subtractive Color TV (p. 12) I. KAMEN 
Video Special-Effects System (p. 14) E. M. 
GORE 

vol. 1, no. 12, Dec. 1950 
Metal-Backed TV Picture Tubes (p. 12) 
C. T. WAUGH 



Society Awards for 1951 



THERE ARE NOW six formal Society Awards 
which are available for presentation an- 
nually to industry engineers who meet the 
qualifications briefly stated here. 

Detailed reports of qualifications of all 
previous recipients of three of the awards 
are presented on pp. 641-643 of the 
JOURNAL for May, 1950, and illustrations 
showing both sides of two of the Society's 
medals appear on pp. 476-477 of the 
April, 1949, JOURNAL. Suggestions for 
possible candidates to be considered by the 
several award committees may be sent to 



Society Headquarters or addressed directly 
to the individual chairmen. 

FELLOW AWARD 

Members in the Active grade who by 
their ". . . proficiency and contributions 
have attained outstanding rank among 
engineers or executives of the motion pic- 
ture industry" may be proposed and con- 
sidered as possible award nominees by the 
Fellow Award Committee. Such proposals 
will be received only from present Fellows 
of the Society and should be addressed to 



257 



Chairman, Earl I. Sponable, Fox Movie- 
tonews, Inc., 460 West 54th Street, New 
York 19, N. Y. 

HONORARY 

The Honorary Membership Award is a 
distinction given to pioneers who have 
contributed inventions of basic importance 
to the industry or whose contributions 
have made possible better production, ad- 
ministration or utilization of motion pic- 
tures. Recommendations for the Honor- 
ary Membership Award may be submitted 
by any member of the Society and must be 
endorsed by at least five Fellows, who are 
required to set forth in writing their 
knowledge of the accomplishments which 
appear to justify presentation of the 
Award. Such recommendations must be 
addressed to the Honorary Membership 
Committee Chairman, Gordon A. Cham- 
bers, Motion Picture Film Dept., Eastman 
Kodak Company, 343 State St., Rochester 
4, N.Y. 

JOURNAL AWARD 

The Journal Award is presented an- 
nually at the Fall Convention of the So- 
ciety to the author of the most outstanding 
paper originally published in the JOURNAL 
of the Society during the preceding calen- 
dar year. Technical merit, originality and 
excellence of presentation are three major 
considerations. The authors of papers of 
nearly equivalent merit often receive 
Honorable Mention. The Journal Award 
Committee, appointed by the President is 
now under the Chairmanship of Mr. Paul 
Arnold, who will shortly be receiving from 
members of his Committee, their recom- 
mendations for the most outstanding 
paper for 1950. His address is: Ansco 
Corp., Binghamton, N.Y. 

SAMUEL L. WARNER MEMORIAL 
AWARD 

Each year the President appoints a 
Samuel L. Warner Memorial Award Com- 
mittee to consider candidates for the 
Award. Preference is given to inventions 
or developments occurring in the last five 
years, and also to inventions or develop- 
ments likely to have the widest and most 
beneficial effect on the quality of repro- 
duced sound and picture. The Award is 



made to an individual and may be based 
upon his contributions of the basic idea 
involved in the particular development 
being considered and also on his contri- 
butions toward the practical working out 
of the idea. The purpose of the Award is 
to encourage the development of new and 
improved methods or apparatus designed 
for sound on film motion pictures, includ- 
ing any step in the process. The present 
Chairman of the Committee is Glenn L. 
Dimmick, RCA Victor Division, Bldg. 
10-4, Camden, N.J. 

PROGRESS MEDAL AWARD 

Written recommendations for candi- 
dates for the Progress Medal Award may 
be submitted by any member of the So- 
ciety, giving in detail the accomplishments 
which appear to justify consideration. 
Qualifications should include invention, 
research, or development which has re- 
sulted in a significant advance in the de- 
velopment of motion picture technology 
and should be seconded in writing by any 
two Fellows or Active members of the So- 
ciety, after which the recommendations 
must be filed with the Chairman of the 
Committee. For 1951, the Chairman is 
D. B. Joy, National Carbon Div., Union 
Carbide and Carbon Co., 30 E. 42 St., 
New York 17. 

DAVID SARNOFF GOLD MEDAL 

Most recent of the Society Awards is the 
David Sarnoff Gold Medal, offered by the 
Radio Corporation of America to a recipi- 
ent to be selected by a committee ap- 
pointed annually by the President of the 
Society. It will be presented annually at 
the Fall Convention to an individual "who 
has done outstanding work in some tech- 
nical phase of the broad field of television 
or in any similar phase of theater television, 
whether in research, development, design, 
manufacture or operation." 

The award will consist of a gold medal, 
a bronze replica and a certificate which 
states the accomplishments which justify 
its presentation. Its objective is "to en- 
courage the development of new techniques, 
new methods and new equipment which 
hold promise for the continued improve- 
ment of television." 



258 



New Products 



Further information about these items can be obtained directly from the addresses 
given. As in the case of technical papers, the Society is not responsible for manufac- 
turers' statements, and publication of news items does not constitute an endorsement. 



The Johnson Kam-Lok is designed to en- 
able a camera to be quickly attached to or 
detached from a tripod. The two parts 
which fit together are released by pulling 
the chain attached to the spring-loaded 
locking phi. The top portion is screwed 
into the tripod bush of the camera and 
can be left there. The lower part is 
screwed onto the tripod. It is suited to 
small movie cameras as well as still 
cameras. The Johnson Kam-Lok is dis- 
tributed by General Photographic Supply 
Co., 136 Charles St., Boston 14, Mass. 




An Underwater exposure meter has been 
developed by Fen John Underwater Photo 
& Equipment Co., 90 Cricket Ave., Ard- 
more, Pa. A Weston Model 852 exposure 
meter has been enclosed in a watertight, 
light-weight, cast-aluminum case. De- 
signed for use in subsurface photography 
and in tropical atmosphere, it weighs in 
air 16 oz and is 3 X 4 X \Y^ in. It is 
priced at $168.00 including Federal tax. 



A Control Track Generator, for syn- 
chronous recording with a tape recorder 
already in use, has been designed by, and 
is available from, Fairchild Recording 
Equipment Corp., 154 St. and 7th Ave., 
Whitestone, N.Y. Called Fairchild Unit 
140, the cost is $335.00 f.o.b. Whitestone. 
It will extend the functions of the tape 
recorders which are performing within the 
specifications published as the NAB 
Primary Standard, so that such recorders 
may meet operational requirements of the 
film and television industries. The con- 



trol track signal is mixed with the program 
signal so that both are simultaneously re- 
corded. When played back on the manu- 
facturer's Pic-Sync reproducer, the re- 
corded program is synchronous with the 
frequency of the power line which supplied 
the original recorder and is therefore in 
"sync" with any other equipment operat- 
ing from the same line at the same time. 



1950 Radiofile Annual is the fifth yearly 
annual now available from the publisher, 
Richard H. Dorf, 255 W. 84 St., New 
York 24. Radiofile is a bimonthly pub- 
lication which indexes and cross-indexes 
by subject all articles of technical interest 
hi 15 leading American radio and television 
magazines and journals, which includes 
the JOURNAL. The sixth Radiofile for a 
year represents all items indexed for that 
year, for the index is cumulative. The 
1950 Radiofile Annual is sold for $0.50, 
and a regular yearly subscription to 
Radiofile is $2.00. 



SMPTE Officers and Committees: The Roster of Society Officers was published 
in the May 1950 JOURNAL. For Administrative Committees see pp. 515-518 
of the April 1950 JOURNAL. The most recent roster of Engineering Com- 
mittees appeared on pp. 337-340 of September 1950 JOURNAL. 



259 




Bruno E. Stechbart 

AFTER 37 YEARS of designing precise mech- 
anisms for the motion picture industries, 
Bruno E. Stechbart resigned from active 
work with the Bell & Howell Co. He 
joined Bell & Howell in 1927 as Develop- 
ment Engineer, became Assistant Chief 
Engineer in 1929 and ten years later, when 
A. S. Howell, one of the Company's found- 
ers and its Technical Director from the 
beginning, retired, he became the Chief 
Engineer. He was elected Vice-President 
in Charge of Engineering in 1944 and held 
that position until July 18, 1950, his 60th 
birthday and the official date of his retire- 
ment. 



His application for membership in the 
Society was dated August 8, 1921 . At that 
time, he was Chief Engineer of the Ameri- 
can Projecting Co. in Chicago. In 1929, 
the Society elevated him to the grade of 
Fellow. 

Mr. Stechbart was born in Lode, Russia, 
(later Poland and Germany), on July 18, 
1890. He came to the United States in 
1906 and, having both interest and apti- 
tude along mechanical lines, found regular 
work in machine shops. To round out his 
knowledge, he studied during off hours and 
at night. In 1913, he became so interested 
in a 35-mm combination camera-projector 
venture that he decided to cast his lot 
permanently with the camera apparatus 
end of the then-growing motion picture 
industry. After a period of experimenta- 
tion, he developed a 35-mm projector, the 
Projectoscope, which was taken up by the 
American Projectoscope Co., a subsidiary 
of the American Film Company. He was 
then employed by them for a period of six 
years. In 1926, he joined the De Vry 
Corporation as Chief Engineer, leaving 
there a year later to become a Development 
Engineer at Bell & Howell. 

Among his former associates, he is 
known for his engineering skill and meticu- 
lous attention to details of design. The 
products developed under his guidance 
present adequate corroboration as does the 
technical article "The Bell & Howell 
Fully Automatic Sound Picture Production 
Printer" that appeared in the JOURNAL for 
October, 1932. He was coauthor with 
A. S. Howell and R. F. Mitchell. At the 
present time, he holds 46 patents and has 
several applications pending. 

Now, in retirement, Mr. and Mrs. Stech- 
bart have moved from Chicago to 206 
49th Street West, R. D. No. 1, Bradenton, 
Fla. 



Meetings of Other Societies 



American Physical Society, Apr. 26-28, Washington, D.C. 
Acoustical Society of America, May 10-12, Washington, D.C. 
American Physical Society, June 14-16, Schenectady, N.y. 
American Physical Society, June 25-28, Vancouver, Canada 
American Institute of Electrical Engineers, June 25-29, Toronto, Canada 
Biological Photographic Association, 21st Annual Meeting, Sept. 12-14, Kenmore Hotel, 

Boston, Mass. 



260 



High-Temperature Film Processing 
Its Effect on Quality 



By Richard Hodgson and Jack Hammer 



The processing of motion picture positive film by high- temperature de- 
velopers and fixing solutions, and subsequent drying by turbulent air, 
results in film quality which is superior in some respects to film processed 
by conventional laboratory methods. Positive pictures, printed from 
an original camera negative, were processed by conventional techniques 
and by high- temperature methods. 



METHODS and techniques to reduce 
the time required for processing 
of motion picture film have been inves- 
tigated for a great many years, but the 
problem has received accelerated atten- 
tion in the last five years. The in- 
creased interest in the problem pri- 
marily comes from three fields : (1) tele- 
vision film recording, (2) military ap- 
plications, and (3) theater television. 

Paramount Pictures' interest has been 
primarily in the rapid processing of 
television film recordings and theater 
television. The first rapid processing 
system was developed by Paramount 
five years ago and it processed and dried 
film in approximately three minutes. 
Continued development effort has re- 



Presented on October 17, 1950, at the 
Society's Convention at Lake Placid, 
N.Y., by Richard Hodgson and Jack 
Hammer, Paramount Pictures Corp., 
Times Square, New York 18. For dis- 
cussion, see the consolidated discussion 
following the next paper in this JOURNAL. 



suited in this time being reduced to 25 
sec by using a high temperature (120 F) 
saturated solutions of caustic developers 
and fixers and turbulent air drying. 

The development work on turbulent 
high-velocity air drying was done jointly 
by Paramount and Raytheon Manu- 
facturing Co. The following paper in 
this JOURNAL describes in detail this 
new technique. Currently theater tele- 
vision equipments which Paramount is 
constructing and selling include this new 
type drier. Figure 1 shows a typical 
installation of the Paramount system, 
including the turbulent air drier. 

In shrinking the complete film proc- 
essing time from the conventional 
commercial laboratory practice of ap- 
proximately 40 min to 1% of that (25 
sec), it was expected that to some de- 
gree the quality would suffer and that 
the print life would be short. This, 
however, has not been the experience 
with the millions of feet of film that has 
been processed in this manner by Para- 
mount. 



March 1951 Journal of the SMPTE Vol. 56 



261 




Fig. 1. The Paramount system, including turbulent air drier. 



To demonstrate these results, two 
prints were made in an identical manner 
from an original camera negative of a 
"Sportlight" short. The same printer 
light setting was used in each case. One 
was processed by the conventional com- 
mercial laboratory procedure in 40 min 
with a 3J/-min carbonate developer at 
68 F. The other print was processed 
and dried hi Paramount's high-speed 
equipment in 25 sec. The gamma of 
both prints is the same. 

(At the Convention, identical sections 
from each print were projected and the 
audience was asked to state which print 
had the higher quality. Approximately 
85% of the members attending the ses- 
sion voted that the print processed in 25 
sec had superior quality.) 



Photomicrographs (445 X) were made 
of identical portions of picture frames 
from both the 40-min (Fig. 2) and 25-sec 
(Fig. 3) prints. Close examination of 
these photomicrographs indicated that 
no detectable difference in grain size 
exists. The grain size is approximately 
1.2 microns in each instance. 

One general film processing technician 
is required to operate the equipment. 
Chemicals are packaged in formula pro- 
portions so that only simple water mix- 
ing is required. This type of equip- 
ment is now in commercial operation as 
part of the theater television installa- 
tions in Detroit, Minneapolis, Toronto, 
Chicago and New York, and additional 
installations are underway. 



262 



March 1951 Journal of the SMPTE Vol. 56 











Fig. 2. Photomicrograph of film processed 40 miii. 




Fig. 3 Photomicrograph of film processed 25 sec. 
Hodgson and Hammer: High-Temperature Processing 263 



Ultrarapid Drying of Motion Picture Film 
by Means of Turbulent Air 

By Leonhard Katz 



The problem of rapid processing of motion picture film has received con- 
siderable attention in the last decade. Although considerable study has 
been made of various methods to increase the speed of developing, fixing 
and washing, relatively little effort has been expended to improve the speed 
of drying. A description is herewith given of a number of different equip- 
ments which were designed after a theoretical investigation had been 
made of the drying problem. These equipments permit the ultrarapid 
drying of motion picture film in a convenient manner with extremely low 
distortion. One application of these units has been incorporated in the 
Paramount Theater-Television Units for the rapid processing of 35-mm 
film. 



Basic Theory of Mass Transfer 

MOTION PICTURE FILM can be con- 
sidered as consisting of a base 
which absorbs practically no water and 
a gelatin layer which absorbs a very 
large amount of water. The drying of 
motion picture film requires that the 
water molecules diffuse out of the gela- 
tin layer to the surface where they can 
be carried off by the surrounding 
medium. The rate of evaporation will 
be completely controlled by the rate of 
mass transfer of the water through the 
gelatin and through the surface layer be- 
tween the gelatin and the surrounding 
medium. 

In order to study this problem more 
closely it is well to consider the general 
theory of mass transfer between a solid 
substance (containing a liquid agent) 



Presented on October 17, 1950 at the 
Society's Convention at Lake Placid, 
N.Y., by Leonhard Katz, Raytheon Manu- 
facturing Co., Waltham 54, Mass. 



and a gaseous surrounding, as shown in 
Fig. 1. From this general theory it will 
then become apparent which param- 
eters can be varied to obtain the maxi- 
mum rate of mass transfer and so ob- 
tain a maximum speed of drying. 

It has long been recognized that in a 
heat and mass transfer problem the 
heat and mass transfer will depend on a 
number of variables as follows : 

1. The area of contact over which 
heat or mass transfer takes place. 

2. The driving force, i.e., the tem- 
perature difference in case of heat trans- 
fer, or pressure difference in case of mass 
transfer. 

3. The resistance to mass or heat 
transfer which is proportional to the 
thickness of the stagnant layer existing 
at the dividing surface between the two 
media. 

The first two variables have usually 
been well recognized and it is the usual 
practice to increase heat or mass trans- I 
fer by increasing the surface area or 1 



264 



March 1951 Journal of the SMPTE Vol. 56 



Symbols, Definitions, Dimensions and Values 



A = 

B = 
C P = 



D 

E a 



f = 

9 = 

h = 

J = 

k = 

L = 

m = 

M = 

n = 

P = 



area normal to heat or mass trans- Pr 

fer, sq ft 

thickness of stagnant layer, ft R 

specific heat = 0.240 Btu/lb. F = R* 

0.514 w/cfm-C for air at 

100 F 

equivalent diameter, D = 4s/p, ft Re 
rate of mass transfer, Ib/hr E 

3.18 X 10 ~ 4 X Jm, for water T 

evaporation 
dimensionless fluid flow friction 

factor; / = 0.046 /R(*- z U 

acceleration of gravity = 32.2 V 

ft /sec 2 on earth W 

wall-to-fluid heat transfer coeffi- 
cient, Btu/hr - sq f t - F a 
film speed through drier, fpm 5 
thermal conductivity = 1.56 X 

10~ 2 Btu/hr-ft 2 -(F/ft) for M 

air at 100 F 
length of film exposed to turbulent p 

air 

mass of water in film sample < 

molecular weight 
number of ducts carrying air 
perimeter of air-carrying duct 



increasing the driving force, i.e., raising 
the temperature. 

The third variable which takes into 
account the resistance to heat or mass 
transfer in the stagnant layer, has not 
always been fully recognized, and it will 
be shown here that a considerable im- 
provement in rate can be obtained by 
reducing the thickness of the stagnant 
layer. This will be the case especially 



= Prandtl number, dimensionless, de- 

fined as 3,600 C p fji/k 
= universal gas constant 
= gas constant per Ib ; for water vapor 

R* = 4.05 X 10~2 ft 3 - atm/ 

R-lb-H 2 O 
= Reynolds number, dimensionless, 

defined as DVp/n = ^w/np 
temperature. T w = average over 

wheel; R = (460 + F) = 

(492 + C X 1.8) 
= air volume, cfm; U = nVs 
= air velocity, fpm 
= mass flow W = 60 t/plb/hr = 



= proportionality constant 

= molecular diffusivity = 0.94 sq ft/ 

hr for water vapor in air, at 100 F 
= viscosity = 1.29 X 10 ~ 5 Ib/ft-sec 

for air at 100 F 
= density = 7.1 X 10 ~ 2 Ib/cu ft for 

air at one atm and 100 F 
= partial pressure, atm; for water 

vapor in air, < = 1.61 X (abso- 

lute humidity) 
time 



MOVING AIR STREAM 



where the stagnant layer is the control- 
ling factor as in the drying of motion 
picture film. 

The stagnant layer can be considered 
as a relatively nonmoving layer of air 
which has the vapor pressure and 
temperature of the liquid on one side 
and the vapor pressure and temperature 
of the gas on the other side. It should 
be recognized, however, that the stag- 
nant layer may not always be the con- 
trolling factor, so that care must be 
taken in applying this theory to other 
drying problems. 

The drying process can be considered 
as a phenomenon normally included in 
the field of molecular diffusion. In 



001 



STAGNANT LAYER 



~-^~^r ~ -7>- LIQUID FILM - 



1 


I 


3 

m 
GAS a > 


_i 
< GAS b 


<r 


5 



Fig. 1. Schematic drawing of the 
theory of mass transfer between a 
solid substance (containing a liquid Fig. 2. Schematic drawing of the gen- 
agent) and a gaseous surrounding. eral theory of diffusion of gases. 



L. Katz: Drying Film by Turbulent Air 



265 



order to get a better understanding of 
the problem, a short review of the theory 
of molecular diffusion is given here. 

General Theory of Diffusion* 

Let us consider two gases, a and 6, as 
shown in Fig. 2, separated by a wall, at 
equal pressure and temperature. If the 
wall is removed, the gases will mix 
rapidly even though no external force is 
applied. It is apparent that an internal 
force of some sort is at work here, and 
the process by which the gases mix is 
called molecular diffusion. It has been 
assumed in elementary theoretical dis- 
cussions that the molecules of the gases 
are extremely hard little balls, that their 
velocity has a Maxwellian distribution, 
and that the mean free path is very 
small. Maxwell assumed that the re- 
sistance to diffusion will be a function 
of the number of molecules in gases a 
and b, the difference in flow velocity, 
and proportional to the length of the dif- 
fusional path. The original equation as 
proposed by Maxwell was: 

rf0 = OLab-^- ~ (V a ~ V b )dx (1) 
lVl a M b 

Under conditions of equal molal diffu- 
sion, and assuming that the Gibbs- 
Dalton law holds and that the gas is a 
perfect gas, it can be shown that: 



E a 
AM a 



RT d<f> a 



<*(0a + 06) 



aM 



% < 2 > 



* Note: Only the results of a rather 
lengthy mathematical treatment are given 
in this paper. For an exact derivation of 
the following equations the reader is re- 
ferred to : ( 1 ) Lecture Notes, Course 2 : 43, 
Prof. J. Kaye, Massachusetts Institute of 
Technology; and (2) "Engineering Report 
on High-Speed Turbulent Air Film Dryer," 
(Private Doc.) Raytheon Manufacturing 
Co., January 19, 1950. The general theory 
of turbulence and its effects have been dis- 
cussed in a large number of papers, most 
of which are listed in the Bibliography. 



where R is the universal gas constant. 
The diffusivity 



(3) 



(0a + 06) 



is now a characteristic which can be 
compared to thermal conductivity in 
heat transfer equations. 

The diffusion of two gases can be 
investigated by making the following 
assumptions : 

1 . Equal molal flux ; 

2. Gibbs-Dalton law holds; 

3. Gas is a perfect gas; 

4. Pressure is constant; and 

5. Temperature is constant. 

For the general case of equal molal 
diffusion of a gas, a, into a gas, 6, and a 
gas, 6, into a gas, a, the following equa- 
tion can be written : 



d 2 



0&) 



(4) 



For the special case of the diffusion of 
a gas, a, through a stagnant layer of 
gas, b, we can assume that the flux of 
gas b is zero and the equation becomes as 
follows : 



R* TB 



J0al 0o2 I 

06m 



(5) 



where R* is the gas constant converted 
for mass of water vapor, 



It should be noted that gas b remains 
stagnant because the molecules of gas a 
tend to carry gas b along in the direction 
of diffusion of a, but the resulting par- 
tial pressure gradient of b causes 6 to 
diffuse in the opposite direction. The 
two rates of travel of gas b are equal and 
opposite and thus cancel. 

For the drying of solids it can initially 
be assumed that the surface of the solid 
is always wet and that the rate of drying 
is entirely governed by the rate of dif- 
fusion from the surface. This is true 
initially until a certain point has been 



266 



March 1951 Journal of the SMPTE Vol. 56 



AIR IN 




FILM IN 



FILM OUT 



Fig. 3a. Experimental setup for film drier. 



reached, whereafter the drying is deter- 
mined by the rate of diffusion of liquid 
through the solid to the surface. The 
mass transfer when the surface is wet is 
governed by: 



E a = A 



R* TB 



06* 



(7) 



For the rate of diffusion from a layer of 
liquid to gaseous surroundings, moving 
in turbulent flow, the following equation 
has been established experimentally by 
Gilliland.* 



~ 
D 



(p r )0.44 



(g) 



It can be seen from Eq. (7) and (8) 
that the fundamental parameter in mass 
transfer is the Reynolds number, a 
dimensionless quantity depending on the 
flow and physical properties of air, and 
on the geometry of the duct through 
which the air passes. We also note 
that this equation contains B, which is 
the effective stagnant air-layer thickness 
through which the water must diffuse. 
If we assume that the rate of evapora- 

* Lecture Notes, Course 2:43, Prof. J. 
Kaye, Massachusetts Institute of Tech- 
nology. 



TURBULENT AIR 



EMULSION SIDE 



Fig. 3b. Cross section of tube 
for turbulent-air film drier. 



tion is not limited by the diffusion of 
liquid through the solid, that is that the 
surface is always wet, an increase in the 
rate of evaporation can obviously be 
obtained by reducing the thickness of 
the stagnant layer, B. This, of course, 
can be done by choosing the parameters 
in Eq. (7) and (8) in such a fashion that 
the Reynolds number is very large and 
the diameter, D, is very small. 

The total rate of evaporation of the 
liquid through the stagnant layer is 
given as follows : 



E 8 (0 2 - 
A " 560 B 



fr ) r L 

R* Ll - 



(9) 



where $1, varying along the length of 
the drier, is the partial pressure of water 
in the air stream, < 2 is at 100% relative 
humidity at the surface of the film, and 
(J) M is the logarithmic mean of </>i and <f>z- 
The final bracket in Eq. (9) is the effect 
of the stagnant air. This bracket can 
be dropped out as long as (f> M <C 1, as is 
the case in most of our applications, 
where a total pressure of 1 atm is main- 
tained. 560 R is used as the tempera- 
ture, T, since 5 was found at 100 F and 
8/T remains nearly constant in the 
operating range. 

It can be seen that the amount of 
water evaporated will be a function of 
the diffusivity of the material, 6, the log 
mean partial pressure difference be- 
tween the water and the air, and the 
thickness of the stagnant layer, B. 

By proper mathematical manipula- 



L. Katz: Drying Film by Turbulent Air 



267 





Fig. 4. Type K-3 turbulent-air 
high-speed film drier. 

tion this can be reduced to the following 
simplified equation: 

E = 2.65 U(<h - <M) 

[1 - e~ 6 - 88 X W-*/BU] (10) 

We notice that this equation contains 
only those variables which are readily 
measurable, i.e., E, U, < 2 , and < u . 
The quantity, B, which is still in this 
equation can be eliminated by combina- 
tion with Eq. (8). By substitution and 
by the introduction of suitable param- 
eters, a proper design can then be 
chosen for a high-speed film drier. For 
example, for the film drier shown in Fig. 
4 substitution of the design parameters 
resulted in the following equation: 



82.6 



(11) 



This equation relates the thickness of 
the stagnant layer to the amount of 
turbulent air blown over the film in 
cubic feet per minute. To determine 
the total amount of water evaporated, 
an equation can be derived from the 
combination of Eq. (10) and (11) which 
results in the following expression: 
E = 2.65 U(<1> 2 - <t> lA )[l-e-*-/U'*\ (12) 



Fig. 5. Rear view of K-3 film drier 
showing turbulent-air heat exchanger 

This equation is the final relation which 
expresses the amount of water evapo- 
rated in a specific length of film as a 
function of the relative humidity and 
the amount of air blown over the film. 
A graphic representation of this equa- 
tion is shown in Fig. 7 as the dotted 
curve which shows the expected amount 
of water removal as a result of the 
application of the foregoing theories. 

This indicates that it should be pos- 
sible to build extremely compact equip- 
ment capable of drying motion picture 
film at high speed and moderate tem- 
peratures (approximately 50 C). The 
dotted curve in Fig. 7 was computed for 
an air flow of 107.5 cfm. Different re- 
sults can be obtained if the air flow is 
also changed and a whole series of curves 
have been obtained showing the rela- 
tionship of air flow, temperature and 
water removed. 
Construction of Equipment 

If motion picture film is to be dried 
rapidly by application of the theories 
previously described, a unit must be 
designed which will permit turbulent 
air to be introduced over the wet film. 



268 



March 1951 Journal of the SMPTE Vol. 56 



The first experimental unit was there- 
fore constructed which consisted of a 
rectangular tube as shown in Fig. 3a, 
with a cross section as shown in Fig. 3b. 
Air at a pressure of 3 psi was introduced 
at the center of the tube at a rate of 
approximately 150 cfm. The film was 
pulled through the tube by a conven- 
tional sprocket drive . Experiments with 
this unit during the early part of 1949 
indicated that even with this elementary 
unit 35-mm film could be dried at a 
rate of 75 fpm, but a number of difficul- 
ties were observed. The film was 
scratched by the bottom of the rec- 
tangular tube and considerable flutter 
existed. Also at this large rate of 
evaporation the film was cooled to near 
the freezing point of water and this 
cooling caused breakage due to reduced 
elasticity of the film. Heating the air 
relieved the situation somewhat but the 
body of the film remained still very cool. 

In order to overcome these difficul- 
ties a second unit was built as shown in 
Figs. 4 and 5. This unit consists of a 
wheel 16 in. in diameter over which the 
film was led, with the back side of the 
film resting on a wheel. Air was intro- 
duced in the rear of the unit and was 
first passed through a turbulent air heat 
exchanger contained in the wheel (see 
Fig. 5). This maintained the wheel at a 
sufficiently high temperature to prevent 
excessive cooling of the film due to 
evaporation. After the air had passed 
through the heat exchanger it then was 
introduced over the film where it passed 
from the top around both sides of the 
wheel to the bottom. A small channel 
in the left-hand bottom side of the wheel 
was used to dry the back of the film by 
forcing air along it to blow off any mois- 
ture droplets still retained on the back 
side of the film. 

This unit operated quite satisfac- 
torily and a very large amount of film 
was dried on this equipment at a rate of 
100 fpm. This unit was also demon- 
strated in conjunction with the Para- 
mount Theater-Television Equipment 




Fig. 6. Experimental setup of Ray- 
theon K-3 film drier with water tank 
removed showing the film-wetting as- 
sembly. 

at a show in September 1949 for the 
Theatre Owners of America in Holly- 
wood, Calif. 

After the demonstration this unit was 
returned to Raytheon for further study 
and an extensive experimental program 
was begun to determine the exact 
characteristics of this unit. A test 
setup was made as shown in Fig. 6 in 



L. Katz: Drying Film by Turbulent Air 



269 




Fig. 7. Graphic representation of the theoretical and computed curves show- 
ing the amount of water evaporated from a specific length of film at given air 
temperatures. 



which measurements were made of all 
the power going into the system, the 
total amount of water evaporated, and 
all the temperatures involved. The 
distribution of pressure and tempera- 
ture was measured at a very large num- 
ber of points along the wheel. As a re- 
sult of these experiments it was found 
that the theoretical curve calculated 
from Eq. (12) and shown in Fig. 7 was 
not matched by the curve obtained in 
actual experiments. 

An investigation of this discrepancy 
showed that the temperature of the film, 
and therefore the temperature of the 
water in the film, was considerably 
lower than the temperature of the air. 
It should be noted that the curve in 
Fig. 7 is plotted as a function of the 



average air temperature and it was be- 
lieved initially that due to the use of the 
highly efficient turbulent air exchanger 
in the rear of the wheel, the wheel and 
film temperature would be equal to the 
air temperature. It was found, how- 
ever, that the temperature difference 
between the air and the film was equal 
to the temperature drop encountered in 
the heat exchanger. Accurate measure- 
ments of the temperature of the wheel 
showed that the rate of evaporation 
followed the predicted theoretical curve 
within experimental errors, if the wheel 
temperature were plotted instead of the 
air temperatures. For instance, in the 
curve of Fig. 7, at an air temperature of 
55 C, the wheel temperature was found 
to be 47 C ; and at an air temperature of 



270 



March 1951 Journal of the SMPTE Vol. 56 



65 C, the wheel temperature was found 
to be 48.5 C. This checks very closely 
with the computed curve. The devia- 
tion from the computed curve, as ob- 
served in the experiments, appears to be, 
therefore, solely a function of the 
temperature difference between the air 
and the wheel. It might be noted here 
that the difference in temperature be- 
tween the wheel and film was very small 
and in the order of not more than 5 C. 




Fig. 8. Type K-4 turbulent-air high- 
speed film drier. 



In order to assure operation of the 
wheel at the exact required tempera- 
ture, a new model film drier (type K-4) 
was built as shown in Fig. 8. In this 
unit heating elements were installed in 
the wheel which were energized through 
slip rings. Approximately 2 kw of 
power was required to maintain the 
drum at the required temperature. The 
air was introduced directly into the slots 
in the rear of the unit and was forced 
over the film in a number of passages 
and exits in the front. The passages 
are formed by a number of movable 
blocks so that the duct width of the 
passage over the film can be varied. 
The diameter of this wheel was reduced 
to 6 in. from 16 in., since the calcula- 
tions indicated that this would be 
sufficient to dry the film. 

Satisfactory results were obtained 



with the K-4 film drier and considerable 
experimental data were obtained drying 
film at 100 fpm. However, an insuffi- 
cient safety factor was allowed to per- 
mit the use of greatly different types of 
film. Up to this point the only film 
that had been used for the experiments 
was Du Pont Film, Type 628B. It 
was found that the amount of moisture 
which could be absorbed in the film dur- 
ing the photographic process varied con- 
siderably with different batches of film, 
with the amount and type of developer 
used, and with the amount of light ex- 
posure on the film. 

In order to perform satisfactory ex- 
periments all tests described above were 
done with undeveloped film which was 
soaked in a wetting tank at constant 
temperature. It was found that only 
under these conditions could reproduc- 
ible results be obtained. 

The final design of a turbulent-air 
film drier which permits a sufficient 
safety factor for the use of almost any 
film of the 35-mm type is shown in Figs. 
9 and 10. This unit (model K-6) incor- 
porates a 14-in. wheel with approxi- 
mately 3 kw of heaters inside the wheel 




Fig. 9. Type K-6 turbulent-air high- 
speed film drier. 



L. Katz: Drying Film by Turbulent Air 



271 




Fig. 10. K-6 film drier installed on Paramount Theater-Television Unit. 



and approximately 2 kw of heat sup- 
plied to the air. A small Standardaire 
blower driven by a 3-hp motor delivers 
approximately 100 cfm at 5 psi. The 
capacity of this unit in water evapo- 
rated is approximately 10 Ib/hr. Figure 
10 shows this unit as mounted on the 
Paramount Theater-Television Unit. 

Distortion 

At the initiation of this project con- 
siderable doubt existed as to the possi- 
bility of survival of the emulsion when 
subjected to high-velocity air streams. 
It was believed that if the air velocity 



became too great the emulsion might be 
stripped off the film or that damage to 
the emulsion might result. 

Although a large number of experi- 
ments have been performed in which air 
velocities have been observed as high as 
550 mph over the film, not a single case 
of damage to the emulsion has been ob- 
served. On the contrary, measurements 
made at the Optical Research Labora- 
tory, Boston University, with film dried 
in the Raytheon K-3 film drier, indi- 
cated that the distortion was consider- 
ably less than the distortion obtained by 
drying film by other means. It is 



272 



March 1951 Journal of the SMPTE Vol. 56 



therefore believed that this method of 
film drying may be especially applicable 
where low distortion is of importance. 
It is believed that the freedom from 
distortion is the direct result of the 
uniformity of application of the turbu- 
lent air and of the rapidity with which 
the water is removed. Therefore all 
film shrinkages are uniform. 

Application of Turbulence to Other 
Diffusion Problems 

The general theory of diffusion can be 
applied to many other problems wher- 
ever the stagnant layer is the controlling 
rate factor in the diffusion problem. A 
number of different equipments have 
therefore been built for various applica- 
tions using turbulent gases and fluids. 
It appears that turbulent air is an 
excellent medium to increase greatly 
the drying speed of most material in web 
form. 

A very important application was 
found in the possibility of increasing 
the speed of photographic development, 
including the actual developing, fixing 
and washing. The speed of the photo- 
graphic development process is greatly 
influenced by the speed of diffusion of 
various liquids into and out of the gela- 
tin. It was found that a considerable 
increase in speed could be obtained by 
passing the developing, fixing and wash- 
ing fluids over the film in turbulent 
form. These experiments will be the 
subject of a later paper. 

Conclusions 

A number of film driers have been de- 
veloped in which turbulent air is used 
as a drying medium. It appears that 
satisfactory results can be obtained with 
such a unit in the ultrarapid drying of 
standard motion picture film in an ex- 
tremely compact space. It was found 
that a wheel 14 in. in diameter will be 
satisfactory for film speeds up to lOOfpm. 
It was also found that the distortion in 
the film with this method of drying is 
very low. 



Acknowledgments 

Part of the work described in this 
paper was done in cooperation with 
Richard Hodgson of Paramount Pic- 
tures, Inc., whose valuable help and 
suggestions contributed greatly to its 
success. The unit 16 in. in diameter, 
shown in Fig. 5, was mounted on the 
Paramount 35-mm theater-television 
equipment and demonstrated in Holly- 
wood for the TOA Convention at the 
Ambassador Hotel in September 1949. 
Approximately 26,000 ft of 35-mm film 
was processed through the 16-in film 
drier at a rate of 90 fpm during this con- 
vention with satisfactory results. 

Part of the work described in this 
paper was done in cooperation with the 
Optical Research Laboratory, Boston 
University, with the assistance of Dr. 
D. E. Macdonald and Dr. R. C. Gunter, 
Jr., from that laboratory. The theo- 
retical derivations described in this 
paper are based on the teachings of 
Prof. J. Kaye, Massachusetts Insti- 
tute of Technology, whose valuable con- 
sulting services greatly helped in the 
speedy solution of many problems. The 
assistance of J. W. Belcher, W. F. 
Esthimer and J. F. Moore, who helped 
in the design of equipment and perform- 
ance of the experiments, is gratefully 
acknowledged. 

. Bibliography 

1. A. R. Burkin, "Ultra-rapid processing 
of photographic materials," Phot. 
Jour., vol. 87B, pp. 108-111, 1947. 

2. F. E. Tuttle and C. H. Green, "Photo- 
graphic race-timing equipment," Jour. 
SMPE, vol. 27, pp. 529-536, Nov. 
1936. 

3. H. A. Miller and J. I. Crabtree, "Com- 
bined developing and fixing solutions," 
Amer. Phot., pp. 76-79, Feb. 1948. 

4. J. I. Crabtree, "Rapid processing of 
films and papers," /. Phot. Soc. Amer., 
vol. 15, pp. 130-136, Feb. 1949. 

5. F. M. Brown, L. L. Blackmer and 
C. J. Kunz, "A system for rapid pro- 
duction of photographic records," J. 
Frank. Inst., vol. 241, pp. 203-212, 
Sept. 1946. 



L. Katz: Drying Film by Turbulent Air 



273 



6. "Dimensional stability of film," Bos- 
ton University Optical Research Lab- 
oratory, Technical Note No. 55, July 
8, 1949. 

7. John C. Johnson, "Measurement of 
the surface temperature of evaporat- 
ing water drops," J. Appl. Phys., vol. 
21, Jan. 1950. 

8. J. Kaye, J. Keenan and William 
Me Adams, "Measurements of friction 
coefficients, recovery factors and heat 
transfer coefficients for supersonic 
flow of air in a pipe," M.I.T. Tech- 
nical Report No. 6418-1, June 1, 1949. 

9. "Dimensional changes of topographic 
film in four drying methods," Boston 
University Optical Research Labora- 
tory, Technical Note No. 10, Jan. 
1948. 

10. B. L. Harris, "Absorption," Ind. Eng. 
Chem., vol. 42, no. 1, pp. 20-24, Jan. 
1950. 

11. Robert L. Pigford, "Absorption and 
humidification," Ind. Eng. Chem., vol. 
42, no. 1, pp. 17-19, Jan. 1950. 

12. Samuel J. Friedman, "Drying," Ind. 
Eng. Chem., vol. 42, no. 1, pp. 38-44, 
Jan. 1950. 

13. Von R. Gran Olsson, "Geschwindig- 
keits-und Temperaturverteilung hinter 
einem Gitter Bei turbulenter Strom- 
ung," Z.A.M.M., vol. 16, pp. 257-274, 
1936. 

14. K. J. Laidler and K. E. Shuler, "The 
kinetics of membrane processes: I. 
The mechanism and the kinetic laws 
for diffusion through membranes; II. 
Theoretical pressure-time relationships 
for permeable membranes," J. Chem. 
Phys., vol. 17, No. 10, pp. 851-865, 
Oct. 1949. 

15. Franz Elias, "The transference of heat 
from a hot plate to an air stream," 
NACA TM-614, Apr. 1931. 

16. E. Forthmann, "Turbulent jet expan- 
sion," NACA TM-789, Mar. 1936. 

17. C. Ferrari, "The transport of vortic- 
ity through fluids in turbulent 
motion," NACA TM-799, July 1936. 

18. H. Reichardt, "Heat transfer through 
turbulent friction layers," NACA 
TM-1047, Sept. 1943. 

19. J. H. Arnold, "Studies in diffusion: I. 
Estimation of diffusivities in gaseous 
systems, Ind. Eng. Chem., vol. 22, p. 
1091, 1930. 



20. B. A. Bachmeteff, Mechanics of Tur- 
bulent Flow, Princeton Press, 1936. 

21. P. Y. Chou, "On an extension of Rey- 
nolds' method of finding apparent 
stress and the nature of turbulence," 
Chinese J. Phys., vol. 4, pp. 1-33, 1940. 

22. S. Corrisin and M. S. Uberoi, "Fur- 
ther experiments on the flow and heat 
transfer in a heated turbulent air jet," 
NACA TM-1865, Apr. 1949. 

23. C. Ferrari, "The transport of vorticity 
through fluids in turbulent motion, 
L'Aerotecnica, vol. 15, pp. 1037-1056, 
1935. Translated as N AC A TM-799, 
July 1936. 

24. J. Jaeger, "Diffusion in turbulent flow 
between parallel planes," Quart. Appl. 
Math., vol. 3, pp. 210-217, 1945. 

25. A. Page, "The behavior of fluids in 
turbulent motion," /. Roy. Aeronaut. 
Soc. London, vol. 37, p. 573, 1933. 

26. E. Forthmann, "Uber turbulente 
Strahlausbreitung," Ingenieur-Archiv., 
vol. 5, pp. 42-54, 1934. Translated 
as NACA TM-789, Mar. 1936. 

27. C. Ferrari, "Sulla Teoria della Turbo- 
lenza," Rendiconti della Reale Acca- 
demia Nazionale dei Lincei, p. 100, 
1934. 

28. H. Reichardt, "A new theory of free 
turbulence," Z.A.M.M., vol. 21, pp. 
257-264, 1941. Translated in J. Roy. 
Aeronaut. Soc. London, vol. 47, pp. 
167-176, 1943. 

29. G. D. Mattioli, "Sopra I'lnfluenza 
della Viscosita del fluido nei Regimi 
Turbolenti," Atti della Reale Acca- 
demia della Scienze di Torino, p. 220, 
1935. 

30. L. Prandtl, "Bermerking zur Theorie 
der freien Turbulenz," Z.A.M.M., vol. 
22, p. 241, 1942. 

31. H. Reichardt, "Gesetzmassigkeiten 
der freien Turbulenz," VDI For- 
schungsheft 414, May-June 1942. 
Reproduced as ARC 6670, FM 586. 

32. H. W. Liepmann, S. Corrisin and J. 
Laufer, "Some measurements in free 
turbulent shear flow," Sixth Interna- 
tional Congress for Applied Mechanics, 
Paris, 1946. 

33. G. D. Mattioli, "Teoria della Turbo- 
lenza," Rendiconti della Reale Ac- 
cademia Nazionale dei Lincei, vol. 17, 
p. 217, 1933. 

34. H. Reichardt, "Heat transfer through 



274 



March 1951 Journal of the SMPTE Vol. 56 



turbulent friction layers," Z.A.M.M., 
vol. 20, pp. 297-328, 1940. Trans- 
lated as NACA TM-1047, Sept. 1943. 

35. G. D. Mattioli, "Dissipazione Verti- 
cose nei Regimi Turbolenti," Rendi- 
conti della Reale Accademia Nazionale 
dei Lincei, p. 87, 1935. 

36. L. Prandtl, "Bericht iiber Unter- 
suchungen zur ausgebildeten Turbu- 
lenz," Z.A.M.M., vol. 5, pp. 136- 
137, 1925. 

37. H. Rouse, "Modern conceptions of the 
mechanics of fluid turbulence," Trans. 
Amer. Soc. Civil Engr., vol. 102, p. 
463, 1937. 

38. T. K. Sherwood, Absorption and Ex- 
traction, McGraw-Hill, New York, 
1937. 

39. L. F. Stutzman, "Mass transfer in 
turbulent flow," doctoral disserta- 
tion, University of Pittsburgh, 1946. 

40. T. K. Sherwood and B. B. Woertz, 
"The role of eddy diffusion in mass 
transfer between phases," Trans. 
Amer. Inst. Chem. Engr., vol. 35, p. 
517, 1939. 

41. G. I. Taylor, "Statistical theory of 
turbulence," Proc. Roy. Soc. London, 
vol. A151, pp. 421-478, 1935. 



42. H. B. Squire, "Reconsideration of the 
theory of free turbulence," RAE 
Farnborough, Report No. Aero 2023, 
Mar. 1945. Abstract in Philosophical 
Magazine, vol. 39, pp. 1-20, Jan. 
1948. 

43. T. K. Sherwood, "Mass transfer and 
friction in turbulent flow," Trans. 
Amer. Inst. Chem. Engr. vol. 36, p. 
817, 1940. 

44. Walton Forstall, Jr., "Material and 
momentum transfer in coaxial gas 
streams," doctoral thesis, Massachu- 
setts Institute of Technology, Course 
II, 1949. 

45. T. von Kdrman, "Some aspects of the 
turbulence problem," Proc. Fourth 
International Congress for Applied 
Mechanics, pp. 54-91, 1934. 

46. K. Victorin, "Untersuchung turbu- 
lenter Mischvorgange," Forschungsar- 
beiten auf dem Gebiete des Ingenieur- 
wesens, vol. 12, p. 16, 1941. 

47. J. S. Hall, A. Mayer, and G. Maslach, 
"A 16-mm rapid film processor," Jour. 
SMPTE, vol. 55, pp. 27-36, July 1950. 

48. C. E. Ives and C. J. Kunz, "Simpli- 
fication of motion picture processing 
methods," Jour. SMPTE, vol. 55, pp. 
3-26, July 1950. 



DISCUSSION 



F. N. GILLETTE: You mentioned that 
the turbulent fluid processing was con- 
siderably faster. Faster than what? 
How does it compare with the spray type 
of processing used in the Paramount 
equipment? 

MR. KATZ: That is a difficult question 
to answer. The developing time in gen- 
eral has been rather vaguely defined. 
In order to get around that problem we 
set up our experiments as follows: The 
system consists of a chamber in which our 
photographic film wedge is suspended and 
we pump liquid developer out of a reser- 
voir through the chamber and back into 
the reservoir again. We measure the 
temperature, pressure and velocity of the 
liquid developer so that we know the 
Reynolds number in the chamber over the 
film. We then compare the speed of de- 
veloping under turbulent conditions with 
the speed of developing required by a simi- 
lar photographic wedge hung in the reser- 
voir under stagnant conditions. There is 



mild agitation in the tank under those 
conditions through liquids being pumped 
around. The ratio between the develop- 
ing time under stagnant conditions and 
the developing time under turbulent con- 
ditions is called the improvement factor. 
It is a little early really to comment on 
this method since we have run only about 
150 films through this turbulent cham- 
ber. The encouraging factor, however, 
is that the improvement appears to be a 
linear function of the Reynolds number 
and we have not found a leveling off of the 
curve anywhere. 

E. W. KELLOGG: [condensed] I want to 
ask Mr. Katz about the static film theory. 
I understand that a number of years ago 
Langmuir, during the development of the 
gas-filled incandescent lamp, concluded 
that in the circulating atmosphere around 
hot filaments there was a stationary film 
whose thickness was cut down by the 
rapidity of the current. Do you know of 
any evidence to prove whether there really 



L. Katz: Drying Film by Turbulent Air 



275 



is a stationary film with a fairly definite 
thickness? 

MR. KATZ: The thickness of the static 
film can actually be determined. It has 
been found that ultrasonic waves will also 
destroy the stagnant layer. 

DR. KELLOGG: How do you know that 
the layer is really stagnant, with a definite 
boundary surface between it and the re- 
gion of turbulent flow? 

MR. KATZ: The only good answer we 
have for this is that the theory has not 
yet been disproved and everything we 
have found so far seems to follow the 
theory accurately. We don't measure the 
stagnant layer directly but indirectly by 
the results which we get, and everything 
that we have done so far lies within experi- 
mental error of our theory. Until the 
theory is disproved we more or less assume 
that we are right. 

ROBERT M. CORBIN: Mr. Katz, I have 
one point I would like to make. In your 
first illustration, you showed film base as 
being more or less an inert material, which 
is far from the fact. In your process, I 
don't think this enters in at all, but if any- 
one should try to dry a film by your 
method after the conventional processing, 
they might get involved in some difficulty 
on account of the amount of moisture 
taken up by the film base. There is very 
little, if any, taken up in the rapid proc- 
essing method; but if you had the film 
base saturated with moisture, and you 
didn't make a very great effort to drive 
the moisture out, at least to the same 
equilibrium at which you are going to keep 
the film, you would get differential drying 
in the roll and a buckling effect which has 
been encountered in the old type of proc- 
essing in the past with disastrous results. 
So, this type of drying should be reserved 
for rapid processing, or at least modified 
to take care of moisture absorption if 
used with other types of processing. 

MR. KATZ: The unit which we have 
built for Boston University under subcon- 
tract for the Air Forces takes 9^-in. wide 
super XX film which has a very heavy 
emulsion. The film is processed by con- 
ventional means with a total processing 
time of approximately 4 hr in which it is 
continuously immersed in liquid. It is 
then dried in the machine at a rate up to 
70 fpm and comes out totally dry. 



MR. CORBIN: My point is concerned 
with motion picture film particularly. If 
you roll it up on the reel, the film is under 
fairly high tension. You don't get drying 
in the center of the roll as rapidly as in the 
outside windings and you can also get 
the outside to dry down more rapidly and 
cause a sort of baggy effect in the center. 
Your edges become shorter than the center, 
with a buckling effect on the screen when 
you project it. There would not be a 
problem with anything like Aero film. 

MR. KATZ: We have measured distor- 
tion rather carefully on Aero film. Very 
fine grid is superimposed on the film and 
then analyzed. We have found no such 
distortion as you mention. 

MR. CORBIN: The distortion I am 
speaking of doesn't take place at the time 
of drying. It takes place during storage 
under rather high tension of rewinding, 
which is generally used for motion picture 
film. It is not a question of distortion at 
the time of drying, but of distortion ef- 
fects after the ultrarapid drying. 

MR. KATZ: We have not run into any- 
thing of that nature yet. It may be 
ahead of us. 

JOHN G. STOTT: I have a couple of 
questions which I think could be tied up 
with one answer. Do you have any data 
on horsepower consumption of drying 
by this method compared to conventional 
methods? 

MR. KATZ: The horsepower rating of 
the unit which is presently installed in the 
Paramount Theater-Television Unit is 
lYz kw. The horsepower rating will al- 
ways go up as the speed of drying increases 
since a certain amount of heat of evapor- 
ation must be supplied. 

MR. STOTT: Do you preheat your air? 

MR. KATZ: Yes, we preheat both our 
air and our wheel. The reason for that is 
that in the evaporation process both the 
air and wheel will cool off and if we don't 
heat the air and the wheel the film will be 
cooled to a point of excessive brittleness 
and break. 

MR. STOTT: I also notice in your paper 
the conspicuous absence of data on rela- 
tive humidity. Does it have any par- 
ticular effect? 

MR. KATZ: The satisfactory operation 
of the unit is independent of relative 
humidity of the air. 



276 



March 1951 Journal of the SMPTE Vol. 56 



MR. STOTT: In other words, you get 
dry film with essentially saturated air. 

MR. KATZ : We have done that, yes. 

H. J. SCHLAFLY: You give figures of 
pressure and air on the first model. Do 
these hold true for the latest design? 

MR. KATZ: No, they vary. The latest 
design uses a thinner opening than the 
first model. The latest design operates 
at 4^-lb pressure and something in the 
vicinity of 80 cfm for the total air con- 
sumption. 

MR. SCHLAFLY: What are the require- 
ments on the blower? 

MR. KATZ: The actual horsepower re- 
quired for the blower is approximately 3 
hp. We have a 5-hp motor installed 
just to be safe. 

MR. STOTT: Mr. Hodgson, with re- 
gard to your demonstration reel, truth- 
fully I was astonished when you said that 
the first reel was processed by conven- 
tional means, whereas the last one was 
processed on your hot machine, because 
I have done some work in this rapid proc- 
essing and have noticed that one of the 
effects (I wouldn't call it a defect) of the 
hot processing is considerable sharpen- 
ing of the toe region of the sensitometric 
curve instead of a long sweeping toe, 
which is characteristic of 5302. It tends 
to sharpen it, which usually shows up as 
more glaring highlights, and I definitely 
thought that the first film, on the basis 
of its highlights, was the rapid processed 
one. Secondly, the rapid processing tends 
to create or produce very much of a blue- 
black tone as compared with the cold 
process, and once again I was fooled on 
that basis, because I thought the first 
film was rapid processed and the second 
one processed conventionally. Do you 
have any comments to make on that? 

RICHARD HODGSON: Not particularly. 
The first portion was done by conven- 
tional 40-min processing. The last sec- 
tion was done in 25 sec. 

MR. STOTT: Do you have any sensi- 
tometry on this? 

MR. HODGSON: No. 

VOICE: I have come to the opposite 
conclusion. Viewed from this distance, 
the second one had definitely more con- 
trast. Also, the second one had more 
blue-black. 

MR. STOTT: I don't think I was at opti- 



mum viewing position down here, but I 
got exactly the opposite reaction from 
yours. How did somebody in the middle 
feel about it? 

VOICE: We couldn't tell the difference. 

MR. HODGSON: Well, that is what we 
like to say. We would like to project 
them side by side too, but the facilities 
here don't lend themselves to this. The 
best we could do was to give you the 
photomicrograph slides in rapid succes- 
sion for what that indicates in grain size. 
It doesn't take care of the other factors 
which you mention, however. 

GEORGE LEWIN: Mr. Hodgson, do you 
happen to know whether this sound track 
was variable density or variable area? 

MR. HODGSON: Area. 

MR. LEWIN : Do you have any data as to 
whether either type of track will respond 
to this treatment in the same way? 

MR. HODGSON: We normally use vari- 
able density. This negative happened to 
be variable-area track and we handled 
it the same way that we handled variable- 
density track. In other words, we de- 
veloped it to a density of 0.2. We gave 
it no special treatment. 

MR. LEWIN: Would you say that you 
can treat the negative just as successfully, 
or just the print? 

MR. HODGSON: We make more nega- 
tives in the course of our recording of tele- 
vision shows than we do positives by about 
a factor of 7 to 1, and they are handled 
by the same method different formulas, 
but the temperatures are the same. 

NORWOOD L. SIMMONS: [Question to 
Mr. Katz] We assume that your remark 
that saturated air can be used must have 
applied to air saturated prior to its heat- 
ing? 

MR. KATZ : That is correct. 

DR. SIMMONS: Secondly, from the front 
row I did note what I thought to be con- 
siderably more density fluctuations in the 
second print nonuniformity, or what 
you might term not mottled, but some 
weaving density fluctuations which ap- 
peared just at the start. Did anyone 
else have such a feeling? 

[Several hands were raised.] 

DR. SIMMONS: There were several others 
who didn't seem to have the courage of 
their convictions when you asked for a 
vote. 



L. Katz: Drying Film by Turbulent Air 



277 



C. R. FORDYCE: I'd like to ask Mr. 
Katz what maximum temperature the 
air reaches in the drying operation, and 
what temperature the film reaches? 

MR. KATZ: There are a number of units 
hi operation at the moment, not all of 
them are alike. In general, we try to 
maintain the film at a temperature of 
approximately 55 C. The air tempera- 
ture varies depending on the design of 
the unit. If there is a heat exchanger in 
the rear it has to enter very much hotter 
to maintain the wheel temperature at the 
desired level of 55 C. In general, we try 
to maintain the air over the film at a tem- 
perature of 70 C. With the air running 
over the film at 70 C, the temperature of 
the film will remain at 55 C. 

J. A. TANNEY: In answer to the question 
about the buckling effect or the in-and- 
out effect that you might get when view- 
ing the projected picture, have you ever 
tried any of those films that have been 
in storage or which have been wound and 
rewound? 

MR. KATZ: As far as the 9^-in. film 
drier is concerned, we have stored 9^- 
in. film for a considerable length of time 
after it has been dried hi the machine, and 
we have not been able to find any diffi- 
culty so far. Maybe we haven't stored 
it long enough or reeled it and unreeled 
it enough. The film in the machine, 
however, gets wound up after consider- 
able tension. 

MR. TANNEY: I was referring to motion 
picture film. I don't suppose you have 
done much of that. 

MR. HODGSON: The first real chance to 
observe this came when the first turbulent 
drier was used in connection with the 
demonstration we made last October in 
Los Angeles for TOA, and that reel or 
series of reels which we dried then we 
have shown to people as recently as last 
August, which was almost a full year later, 
and the projectionists have reported no 
trouble hi running it through. It seems 
to handle just as a normal reel. 

DR. KELLOGG: One more question 
occurs to me. What provisions do you 
make for keeping your air clean? It 
seems to me that this violent blowing 
would tend to deposit dust on your film 
unless some rather extraordinary filtering 
precautions were taken, or else re-using 



the air after precipitating out the moisture. 
MR. KATZ : We have had that problem, 
especially at the point where the air is 
first introduced on the film. At that point 
the air has to go around a corner and any 
particles in the air would hit the film 
straight on. In our 9^-in. wide film 
drier we found that pitting of the film 
would occur if we purposely blew sand into 
the machine. Sufficient filtering, how- 
ever, was obtained by means of an ordi- 
nary air filter similar to the one used in an 
automobile engine. 

Supplementary Discussion* 

The stagnant-layer theory is specifi- 
cally based on the principle that the 
speed of diffusion of the liquids through 
the gelatin or other solids is sufficiently 
great so that the surface can always be 
assumed to be wet. If the surface of the 
gelatin is wet then the only resistance to 
diffusion will be posed by the thickness of 
the stagnant layer. At low tempera- 
tures the driving force is relatively small 
and the entire resistance to evaporation 
is governed by the thickness of the 
stagnant layer. The rate of evapora- 
tion will still be relatively small at low 
temperatures even if the stagnant layer 
is reduced because the driving force is 
relatively small. Of course, reduction 
in the stagnant layer will bring about a 
proportional increase in the rate of 
evaporation. Consequently, if, for in- 
stance, at low temperature the normal 
drying time for one section of film were 1 
hr, and the stagnant layer were reduced 
to ^jo of its thickness by the applica- 
tion of turbulent air, then the drying 
time would be reduced to 2 min. 

Inasmuch as the rate of evaporation is 
still relatively low, the rate of diffusion 
of the liquid through the gelatin is so 
small that the surface can be assumed to 
be wet at all times. If, however, the 
temperature is now raised, which in- 
creases the driving force, the rate of 
evaporation will be increased to a point 

* Communicated by the author on Decem- 
ber 14, 1950, in order to clarify points 
raised during the discussion. 



278 



March 1951 Journal of the SMPTE Vol. 56 



where the diffusion through the gelatin 
itself will become the limiting factor, 
i.e., a point may be reached in which the 
surface layer of the gelatin closest to the 
gas side is relatively dry, whereas the 
bottom of the gelatin closest to the base 
of the film would be relatively wet. 

The problem is then resolved into a 
transient diffusion problem in which a 
humidity gradient is set up through the 
gelatin. This humidity gradient will 
only become important when the rate of 
evaporation has been greatly increased. 
It was found experimentally that this 
limit becomes apparent when positive- 
type stock is dried in a tune shorter than 
1 sec or when negative stock is dried in a 
time shorter than 2 sec. In general, 
therefore, film driers must be designed 
in such a fashion that the emulsion of 
the film is exposed to turbulent air for a 
period of not less than 2 sec. The 
actual film speed that can be obtained 
can, of course, be made as large as one 
wishes by simply increasing the time 
during which the film is continuously 
exposed to turbulent air. 

With reference to the extremely low 
temperatures realized in the use of the 
first model, it can be stated that this was 
not originally anticipated. It was 
hoped that as a result of the turbulence 
the heat transfer between the air and 
the film would be sufficient to maintain 
the film at room temperature. How- 
ever, it was found that approximately 
40% of the heat was supplied from the 
air and 60% of the heat had to be sup- 
plied by direct conduction or radiation 
into the film itself. Consequently, the 
heating of the back side of the film was 
necessary to maintain the film at room 
temperature or higher. 

With reference to the use of raw film 
in the experiments, I would like to ex- 
plain that a large number of experi- 
ments were first made with processed 
film which had been run through Para- 
mount's 35-mm machine. It was found, 
however, that small changes in the de- 
veloper or hypo concentrations plus 



small variations in the thickness and 
quality of the emulsion itself produced 
variations in water absorption of the 
film of approximately 60%. As a re- 
sult it was not possible to use processed 
film in the experiments, as consistent 
data could not be obtained. A number 
of experiments were then performed in 
which the total water absorption of the 
processed film was checked carefully and 
the total spread in the water absorption 
figures was carefully noted. The ex- 
perimental determination of the rate of 
evaporation at various temperatures and 
Reynolds numbers of the air was then 
performed on raw stock which again had 
been carefully measured as to its water 
absorption qualities. It was found 
that the amount of water absorbed in 
the raw stock was to a great extent deter- 
mined by the temperature of the water 
in which it was immersed. During the 
experiments the temperature of the 
water was maintained at a point where 
the water absorption of the raw stock 
was approximately equal to the water 
absorption of the processed stock. 
Adequate corrections were then made in 
the design of the film drier to produce a 
safety factor which would compensate 
for the spread in the water absorption 
qualities of the processed stock, so that 
even under the worst conditions proc- 
essed stock could be dried in the re- 
quired time. Spot checks were made at 
various points during the experiments 
with the processed film to compare the 
result between processed film and raw 
stock at a number of points in the curves. 
With reference to the type of film dis- 
tortion which was measured and the 
methods used for measurement, I would 
like to defer the discussion of this to a 
later paper which is now in preparation 
in cooperation with Dr. R. C. Gunter of 
the Optical Research Laboratory, Bos- 
ton University. Also, additional de- 
tails concerning the actual design of film 
driers and general design charts will be 
presented in that paper. 



L. Katz: Drying Film by Turbulent Air 



279 



Television Transmission 

in Local Telephone Exchange Areas 

By L. W. Morrison 



The functions of a video transmission system in a local exchange area in 
providing mobility for the pickup camera and interconnection with the 
intercity networks are discussed; and an analysis of some of the television 
transmission problems is presented. A description is given of the physical 
and electrical characteristics of the various types of cable facilities, the 
video amplifiers, and equalizers now employed; and an example of the tele- 
vision transmission performance obtained is included. 



Hf^HE TELEVISION transmission dis- 
-- tribution system in the local ex- 
change area fulfills a most necessary 
function in the over-all television broad- 
casting system. The sole purpose of 
network broadcasting is to attract the 
greatest possible audience. The local 
video distribution system, as the means 
whereby the initial and final connection 
is accomplished, contributes to the 
efficiency of the network pattern in two 
ways: first, it allows greater mobility 
for the pickup camera, thus permitting 
a wider range of subjects and more 
attractive program material to be 
offered to the public; and second, it 
furnishes the means by which the pro- 
gram may be made available to viewers 
far outside the present restricted cover- 
age of a television broadcast trans- 
mitter. In the future, it is believed, the 
local television system may in a similar 
fashion contribute to efficient distribu- 



Presented on October 15, 1950, at the 
Society's Convention at Lake Placid, 
N.Y., by L. W. Morrison, Bell Telephone 
Laboratories, Murray Hill, N.J. 



tion of televised material for theater 
consumption. 

At present, the local video system is 
called upon to furnish the connection 
between the remote pickup camera and 
the master control area of the local 
studio, the connection between the 
studio and its local broadcast trans- 
mitter location, and the connection be- 
tween the studio and the intercity net- 
work facility. 

It is apparent that we are here dealing 
with a transmission facility which has 
two unique characteristics. First, we 
are concerned with relatively short 
distances, say 1000 ft to 20 miles; and 
second, the system is entirely a point-to- 
point transmission medium as con- 
trasted with area coverage of the usual 
broadcast transmitter. In addition, 
the very nature of the function dictates 
that privacy of the connection be as- 
sured. 

The Bell System presently employs 
both wire and microwave television 
transmission methods in the local tele- 
phone exchange area. The choice of:.' 
medium here, as always, is one of| 



280 



March 1951 Journal of the SMPTE Vol. 56 



economics. For example, in the Los 
Angeles, Calif., area all television broad- 
cast transmitters are located on a 
mountain range overlooking the metro- 
politan area, approximately 20 miles 
from the Hollywood location of most 
studios. The terrain is such that 
microwave methods have an over- 
whelming advantage over wire circuits 
for the main circuit, though even here 
the final connections to the various 
broadcast transmitters on the moun- 
tain are made by wire circuits just as 
the studio-to-Hollywood Central Office 
pickup connection is made. In many 
other metropolitan locations the micro- 
wave line-of-sight requirement is diffi- 
cult to achieve and in these cases wire 
television facilities are almost ex- 
clusively employed. In our opinion, 
wire and microwave television trans- 
mission techniques will tend to comple- 
ment rather than compete in the service 
of providing a sound economic local 
point-to-point television connection. 

The following discussion will be con- 
fined to the local telephone exchange 
area television transmission systems 
employing wire as the transmission 
medium. It is proposed to analyze 
the problem considering the terminal 
conditions, the characteristics of the 
cable plant, and the gain and equaliza- 
tion required. A description of the 
equipment now in use will be followed 
by a discussion of typical performance 
obtained. 

Terminal Conditions 

The television studio equipment 
which represents one terminal condition 
for a local video interconnection is 
generally operated on a 75-ohm un- 
balanced-to-ground impedance basis. 
This is a matter of economy of com- 
ponents and within the restricted con- 
fines of a studio area is permissible since 
the signal levels are relatively high and 
the transmission distances involved are 
small. 

In the telephone plant where greater 



lengths of exposure to interfering elec- 
trical fields are the rule, the balanced- 
to-ground impedance is universally 
employed. The characteristic im- 
pedances of telephone cables which are 
available for video transmission pur- 
poses range from 90 ohms to about 140 
ohms. The equipment design at present 
is based on a balanced impedance of 110 
ohms. 

The signal level available at the 
studio for transmission over a local 
interconnecting circuit has been gen- 
erally of the order of 2 v peak-to-peak 
and the input level required at a tele- 
vision broadcast transmitter or at the 
input to the studio master control equip- 
ment has varied from 0.25 to 1 v peak- 
to-peak. 

The signal level which is impressed on 
the telephone cable is maintained at 
the highest possible economic value to 
insure the best signal-to-noise per- 
formance. At present the common sig- 
nal amplitude at the input terminals 
of a cable is generally of the order of 2 v 
peak-to-peak. In the television operat- 
ing centers which represent the ter- 
minals of the intercity network a 
standard level of 1 v peak-to-peak 
across 110 ohms has been adopted. 

The local video transmission equip- 
ment then must provide for the con- 
version of the signal from the form 
available at the terminal ends to one 
suitable for transmission over the tele- 
phone cable. This involves treatment 
of both impedance and signal levels. 

Telephone Cable Characteristics 

The character of the transmission 
medium exerts the primary influence on 
the design of the bulk of the equipment 
employed in local video interconnecting 
circuits. The characteristics involving 
impedance, attenuation versus fre- 
quency, and the shielding, as well as 
the pair location with respect to sources 
of induced unwanted interference, must 
be considered. The impedance and 
attenuation characteristic dictate the 



L. W. Morrison: Local TV Transmission 



281 




40 60 80100 200 400 600 1000 2000 
FREQUENCY IN KILOCYCLES PER SECOND 



4000 



Fig. 1. Attenuation of paper-insulated cables at video frequencies. 



type of termination to be employed and 
the degree of equalization required. 
The shielding properties of the cable, 
together with a knowledge of the degree 
of exposure to sources of interference, 
finally set the signal-to-noise perform- 
ance of the circuit. 

A wide variety of paper-insulated 
interoffice and subscriber cables exist 
in the telephone plant. Gauges rang- 
ing from No. 10 to No. 26 are employed 
with many available routes consisting 
of a composite of these. Figure 1 shows 
the attenuation-frequency character- 
istics for a number of cables which have 
been used for video purposes in the past. 
The loss at 4 me (megacycles), which 
must be compensated for by equaliza- 
tion and gain, varies from about 25 to 
95 db/mile for the types shown. 

These paper-insulated pairs are ad- 
jacent to others, perhaps as many as 
1800 in a single lead sheath, and are 
therefore subject to induced interference 
from signals in these associated message 
circuits. Impulse-type noise is com- 
monly transmitted over a message cir- 
cuit resulting in large measure from the 
complex switching operations of the 
modern telephone plant. In addition, 



low-frequency signals may be induced 
into the video pair from neighboring 
power circuits. 

The minimum level to which the 
video signal may be attenuated involves 
the total interference level at hand and 
a signal-to-noise performance objective. 
With this information and a knowledge 
of the maximum undistorted power out- 
put of the video repeater or the trans- 
mitting terminal, the maximum usable 
gain required is determined. Then by 
a consideration of the attenuation 
characteristic of the cable, the repeater 
spacing may be computed. 

For the paper-insulated pairs under 
consideration, impulse-type noise gen- 
erally is controlling, and repeater gains 
of about 65 db result. For No. 26 
gauge this will result in a repeater spac- 
ing of about 0.7 mile. 

There is finally another restriction in 
the maximum gain which can be used 
in a video repeater. The coupling be- 
tween the input and output of the re- 
peater must be limited if unwanted 
regeneration effects are to be minimized. 
This coupling from output to input 
terminals via adjacent pairs limits the 
gain usually to the order of 40 db. This 



March 1951 Journal of the SMPTE Vol. 56 



* ( L* 1< NO. !6 CONDUCTOR ~ 

U ! / 1 i 

! yk I/ ^W POLYETHVLENE 3^ 

\ 1 \ I , TAPE 



NO- 18 CONDUCTOR- *4 



O.CO3 OUTER 
SPSRAL 
COPPER 




OLYETHYtEN6 
STRfNG 





OUTER 
BSAiO 
COPPER 



16PSV 
SPfRAL SHIELD 



0-003 
SPiRAL 
COPPER 



16 PSV 
LOHSiTUDSNAL SHIELD 



754. C 
DOUBLE BRAID SHIELD 



Fig. 2. Construction of shielded video cables for local exchange area use. 



situation may be remedied by the use of 
separate cable sheaths at the input and 
output of each repeater. 

At the present time the majority of 
video interconnecting circuits in the 
local exchange area employ a special 
shielded pair for transmission, thus re- 
ducing materially the restrictions pres- 
ent with paper-insulated telephone 
cables just discussed. Figure 2 illus- 
trates the mechanical features of con- 
struction of some of these shielded tele- 
vision pairs. 

The 16 PSV (polyethylene shielded 
video) pair uses polyethylene string 
and tape to support the 16-gauge con- 
ductors within a copper shielding tube. 
As shown, both spiral copper tape and 
a combination of spirally wound tape 
and a longitudinal seamed tube have 
been employed. Such a conductor is 
composited directly into an interoffice 
cable replacing about 20 of the paper- 
insulated pairs. A number of such 16 



PSV conductors may be placed within a 
common lead sheath, the improved 
shielding allowing their use simultane- 
ously in either direction for trans- 
mission. 

For interior wiring within the central 
office and at the terminal ends of the 
circuit, as well as for temporary short 
outside routes, the double-braided flexi- 
ble cable shown in Fig. 2 has been ex- 
tensively employed. This cable has a 
solid polyethylene extruded core con- 
taining two No. 18 conductors. 

The attenuation-frequency character- 
istic of the polyethylene dielectric type 
of video cables is given in Fig. 3. 

When the shielded video cable is used 
the impulse-noise level is greatly at- 
tenuated and the allowable repeater 
gains may be increased. With an 
attenuation of 17 db/mile, as indicated 
in Fig. 3, it allows repeater spacings 
upward of 5 miles to be employed. 

To provide adequate transmission 



L. W. Morrison: Local TV Transmission 



283 



performance for television signals it is 
necessary to equalize the attenuation- 
frequency characteristic of the cable. 
This is accomplished by furnishing the 
gain required in the form of video ampli- 
fiers, having a constant gain-versus-fre- 
quency characteristic. To allow the 
equalization of any length of circuit and 
of circuits consisting of several gauges 
of conductors a variable plan for equali- 
zation is used. Fixed and variable units 
of equalizers are provided in a fashion 
allowing adequate compensation for 
any specific cable situation. 

Predistortion of the television signal 
may be employed in certain instances 
to improve the signal-to-noise per- 
formance of the video system. The 
method involves the amplification of the 
higher frequency components of the 
video signal prior to transmission over 
the cable. At the receiving terminal a 
; restorer network having a loss-frequency 
characteristic complementary to that of 
the predistorter is introduced. The 
signal is unaffected by transmission 
through these two networks while noise 
introduced into the cable circuit is 
attenuated by the restoring network. 
Figure 4 indicates the characteristics of 
the predistorter and restorer networks. 
An impulse-noise improvement of about 
16 db is realized in practice. 

Equipment Description 

There are three basic equipment 
units which are required in a local wire 
television transmission installation. 
These include a transmitting terminal, 
a repeater, and the receiving terminal. 
The primary function of the trans- 
mitting and receiving terminals is one 
of conversion of the signal from the 
customer's equipment to a form suit- 
able to the cable pair in the telephone 
plant. The repeater unit furnishes the 
bulk of the gain required to overcome 
the transmission loss of the cable and 
the means to properly equalize its loss- 
frequency characteristic. 

The Video Repeater. The repeater 



arrangement is shown as a block sche- 
matic in Fig. 5. It consists of four 
functional components : an input ampli- 
fier, an equalizer, an output amplifier 
and necessary power supplies. 

The input amplifier has as its func- 
tion the amplification of the signal as it 
appears across the cable terminals to a 
level sufficiently high to permit equaliza- 
tion. It consists of an input coupling 
network, a balanced input stage and a 
cathode follower output circuit. The 
coupling network has a gain-versus- 
frequency characteristic which increases 
to about 13 db at 4 me. This character- 
istic in conjunction with one of the 
equalizer units is capable of compensat- 
ing for the loss-frequency character- 
istic of about 1 J^ miles of 16 PSV cable. 

The schematic details of the balanced 
input stage are shown in Fig. 6. It is 
arranged to provide a large amount of 
feedback to longitudinally induced noise 
voltages which arrive at the repeater. 
The longitudinal suppression here ob- 
tained is of the order of 75 db at low 
frequencies, the region of the most 
troublesome noise conditions. In addi- 
tion, impedance balancing controls are 
provided to further minimize these 
effects. The voltage gain of this input 
amplifier is about 14 db and the output 
circuit is arranged to connect directly 
to the following equalizers on a 1000- 
ohm impedance basis. 

The equalizer portion of the video 
repeater consists of three types of units : 
a basic equalizer whose loss-frequency 
characteristic in conjunction with the 
input amplifier coupling network will 
compensate for about 25 db of 16 PSV 
cable loss at 4 me ; a variable unit which 
will equalize 10 db of cable loss in 1-db 
steps and the fixed equalizer which will 
equalize a 10-db section of 16 PSV cable. 

By using one or more amplifier cou- 
pling networks, and basic, variable and 
fixed equalizers, we have the means to 
provide equalization of any specific 
length of cable limited finally by the 
gain capabilities of the repeater. 



284 



March 1951 Journal of the SMPTE Vol. 56 



28 



24 



20 



16 



? 12 



A. 16 PSV SPIRAL SHIELD 

B. 16 PSV LONGITUDINAL SHIELD 

C. 754 C DOUBLE BRAID SHIELD 




1 



f 



I 



4 6 8 10 20 40 60 100 200 400 1000 
FREQUENCY IN KILOCYCLES PER SECOND 



4000 



Fig. 3. Attenuation of shielded video cables. 




200 300 400 600 8001000 2000 3000 

FREQUENCY IN KILOCYCLES PER SECOND 



5000 



Fig. 4. Loss-frequency characteristics for predistorter 
and restorer networks. 



L. W. Morrison: Local TV Transmission 



285 



50 



Ul 
ZZ 

i 1 o-H 

A w\ 

A A A 


I L 

=] C 

=1 1= 


u u 

NOTE 

EQUALIZERS AS REQU 
F INPUT CABLE PAIR: 

>E EQUALIZERS 


$ 

o J 


'VVV 


1 


H* ~m 3 


1 






pg oo. ^l 

0- > ujO 
^" Z ' 




QC 


1 1 


z 




D^ 


A - 






| 


/\ - 






OS 


1 


^ 


p 




! 

a. 


> c 

? 


> 

in 


SP 


< 


(C 





H .z, 








i^< 

s 


|. 


or 

Ul 
U. 


UJ-I 

5- 

<D 




5<0 


O 
Ul 
(C 




iu a 


1 




? 


II 






<O 
> 


a ^2 


1 




^ 


Di 


. A 






ZCL 


Z_A - 






5 > 


i i 














D 


1 I 


a 

LU 








1 

Q.D 




AxAA 


1 



III 













t 

I I 



w n $ o <o 



-0 = 

>o 



286 



March 1951 Journal of the SMPTE Vol. 56 



V3 
6AG7, 



+300 
VOLTS 




DISCONNECTED 

WHEN NETWORK 

IS USED 



+300 
= VOLTS I 

Fig. 6. Video repeater input amplifier simplified schematic. 



The output amplifier shown sche- 
matically in Fig. 7 provides 30 db of 
gain to the signal after it has been 
equalized and normally operates to 
furnish a 2-v peak-to-peak signal to the 
following cable section. Through the 
use of a network similar to that dis- 
cussed in connection with the input 
amplifier a pre-equalized signal may be 
impressed on the following cable section 
with an attendant improvement in 
signal-to-noise performance. Balanced 
or unbalanced resistive terminations are 
also available if desired. 

The physical arrangement of a video 
repeater is shown in Fig. 8. In this 
case a single repeater, cabinet-mounted 
for semipermanent use, is shown. In 
the central office three such repeaters 
are mounted on a standard 11-ft relay 
rack. A central office installation is 
shown in Fig. 9. 

The Transmitting Terminal. Two 
versions of the video transmitting 
terminal are shown in Figs. lOa and 
lOb. The most commonly employed 
arrangement is given in Fig. lOa and 
consists essentially of a video repeating 



coil which permits coupling the usual 
75-ohm unbalanced studio equipment to 
the balanced telephone cable, and a pre- 
distorting network which is employed 
when abnormally long cable circuits are 
involved. The repeating coil has a 
substantially flat transmission char- 
acteristic over the video band with the 
3-db loss points at 30 cycles/sec and 
at about 8 me. The physical arrange- 
ment of this type of video transmitting 
terminal is shown in Fig. 11. 

In a few instances the first section of 
cable may be of considerable length. 
Here an amplifier may be employed as 
indicated in Fig. lOb. This amplifier 
is identical with the output type em- 
ployed in the video repeater just de- 
scribed. The use of this alternative 
transmitting terminal permits the first 
cable section to extend to 33/ miles in 
length as compared with a maximum 
length of 1.7 miles when a repeating- 
coil type of transmitting terminal is 
employed. 

Receiving Terminal. In Figs. 12a and 
12b two alternative receiving terminal 
arrangements are shown, the choice 



L. W. Morrison : Local TV Transmission 



287 



(M013Q) 

O o g'v 01 



Q30NV1VQNn 
USA OM1 bO 3 NO Ol 




^K 




I A A A 




VXA/ ^5^^- 

UOOOl \ 



USA 



bO 

UOU Ol 






n" 


H 
o 

HI 


Q 

O ^ 





z 


<2 0< 


G 


z 


s tfi 


> s 


8 


A A A 


A A A 


-^ ^ 






O 

TLi 


'I 




^T5 


N 

A .AA. i 


h \AA/ 


_1 ^ 


V V v^ 

c 


a 


5 






R 


i 


?5 






1*7 




/~ 


(D 



WObd 



blVd 31SVO 9NIO910O Ol 





288 



March 1951 Journal of the SMPTE Vol. 56 




= 



- 
a 

s 



CO 

si. 



L. W. Morrison: Local TV Transmission 



289 




-J 




11 

1! 



1 



If 



290 



March 1951 Journal of the SMPTE Vol. 56 




ACK STRIP 



REPEATING COIL 



- DISTORTER 



Fig. 11. Video transmitting terminal in wall-mounted bracket. 



being primarily dependent on the loss 
of the final cable section. For shorter 
lengths of cable where the loss at 4 me 
does not exceed 27 db, the arrangement 
of Fig. 12a is commonly employed. 
This terminal consists of a video repeat- 
ing coil, a restorer network if required, 
and clamper amplifier. The clamper 
amplifier whose operation has been dis- 
cussed elsewhere, 1 affords protection 
against low-frequency noise interfer- 
ence, as well as minimizing the trans- 
mission distortion introduced at very 
low frequencies by the repeating coil. 
The clamper amplifier here employed 
has a gain of 18 db and will furnish a 
2-v peak-to-peak video signal into a 
75-ohm impedance for use by the 
customer. It affords a 31-db reduction 
of 60-cycle interference. 

If the final section of cable is greater 
than about 1^ miles, the receiving 
terminal arrangement shown in Fig. 12b 
is used. Here the basic components of 
the video repeater are employed and, if 
required, in addition a clamper ampli- 
fier is included as shown. 

System Performance 

The objective in the transmission and 
equipment design of the video system 
just described is to provide adequate 

1 S. Doba, Jr., and J. W. Rieke, "Clampers 
in video transmission," AIEE Trans., 
vol. 69, Pt. 1, pp. 477-487, 1950. 



transmission of video signals in local 
exchange areas with a minimum of cost. 
The final cost of such service is de- 
pendent only in part on the initial 
equipment and cable investment. The 
installation and maintenance charges 
contribute substantially to the over-all 
system cost. The ability to provide 
prompt and straightforward intercon- 
nections in the local area, regardless of 
the specific peculiarities of the particu- 
lar situation, is a basic requirement of 
such system design. 

The following two circuit examples 
are given to illustrate the application of 
these video components to the specific 
conditions encountered. 

In Fig. 13 the transmission per- 
formance of a 2.78-mile circuit contain- 
ing one repeater and simple terminals is 
given. 

In Fig. 14 a longer and more complex 
example is presented. Here the total 
circuit length is 11.72 miles and is 
further complicated by the fact that one 
of the cable sections is 5 miles in length. 
This extreme condition is met here by 
transmitting the signal at a somewhat 
higher-than-normal amplitude over this 
section and also by operating the follow- 
ing repeater at somewhat increased gain. 
In this instance the noise conditions 
were such that this procedure was 
allowable. A mop-up equalizer is also 
employed to reduce the residual trans- 



L. W. Morrison : Local TV Transmission 



291 




I 



II 

.- o 

I S 



3 




Sc 

s, 



v 



~rl 

> > ^ > > 



)oC=l 1=1 



292 



March 1951 Journal of the SMPTE Vol. 56 



TRANSMITTER 
TERMINAL 


REPEATER IN 
CENTRAL OFFICE 


RECEIVING 
TERMINAL 


r ' i 




INPUT I ojjo J L 


*> 




cjjo i 


1 1 










CIR< 


:UIT MILES .^i, 


-V2t, S c AMPLIF7B 



OUTPUT 



CIRCUIT 


MILES 


4-MC LOSS 
IN DECIBELS 


Li 


1.67 


30 


L2 


1.11 


20 


TOTAL 


2.78 


50 



RESPONSE IN DECIBELS 
iiit 

* UJ fO O M 








































=c^r . 





=; 


. 


! * 


.**- 




>-^ 


== 


- 


^-_ 


_^ 


** 


^x 


=^x 


^; 






































\ 












































, 






i 


i 




i 




i 


. 








. 


t 



8 10 20 40 60 80 10O 20O 400 600 1000 

FREQUENCY IN KILOCYCLES PER SECOND 



2000 



6000 



Fig. 13. Video circuit application 2. 78-mile circuit. 



TRANSMITTER 
TERMINAL 




CIRCUIT 


MILES 


4-MC LOSS 
IN DECIBELS 


LI 


1.67 


30 


L 2 


5.0 


90 


1-3 


1.33 


24 


LA 


2.39 


43 


L 5 


1.33 


24 


TOTAL 


11.72 


211 



M = MOPUP EQUALIZER 

R = RESTORER 

C = CLAMPER AMPLIFIER 




: 4 6 8 10 20 40 60 80100 200 400 600 1000 2000 

FREQUENCY IN KILOCYCLES PER SECOND 

Fig. 14. Video circuit application 11.72-mile circuit. 
L. W. Morrison: Local TV Transmission 



293 



mission distortion occasioned by the 
large number of components in tandem. 
Acknowledgment. The author is in- 
debted to C. N. Nebel of the Bell Tele- 
phone Laboratories for the illustrations 
used, as well as for the performance 
data for the two local video circuits 
discussed. 

Discussion 

E. W. KELLOGG: How does the attenua- 
tion characteristic of the cable you were 
showing in this picture compare with that 
of your long-distance coaxial cable, and to 
what is the difference principally due? 

MB. MORRISON: In the case of the 16 
PSV (polyethylene shielded video) cable, 
the characteristics are almost identical. 
As to attenuation versus frequency: 
essentially, the attenuation rises almost as 
the square root of the frequency in all 
these cables where we have no high dielec- 
tric loss in the region we are talking about. 
Now, in the case of the coaxial cable, we 
have polyethylene dielectric and here we 
have the same, so we do have the same 
form of characteristic. And as to the 
amount of attenuation, it depends upon 
the copper. In the case of the coaxial, we 
are able to put more copper in the same 
size of tube, if you will. The attenuation 
is lower per mile. The unbalanced charac- 
teristic of the coaxial, however, does not 
permit its use for video transmission over 
very long distances. The coaxial is not 
well shielded for low-frequency power 
interference. It has a steel tape, to be 
sure, as well as copper; but it is not ade- 
quate for television transmission. We are 
able to transmit video frequencies on 
balanced conductors in these places where 
we have high levels of interference; and 
that is generally the case in any metro- 
politan area where these cables are run. 
So, in the coaxial we limit the lower fre- 
quency of transmission. On the present 
LI system we limit it to 60 kc. For tele- 
vision we limit it to about 200 kc. The 
top frequencies can be made anything if 
you are willing to put in enough gain and 
narrow the repeater spacings. The wider 
the band, the closer the spacings for the 
amplifier. 



DR. KELLOGG: Is the polyethylene di- 
electric continuous or is it merely a series 
of spacers? 

MR. MORRISON : The dielectric does not 
occupy all of the space. It is not solid. In 
the coaxial system we use beads, at ap- 
proximately 1-in. spacing, little discs of 
polyethylene; and in the case of this video 
cable which you saw, we used string spi- 
rally wound around it and tape over that, 
so that a good percentage of the volume is 
air. That is just a matter of handling, 
and of course the material itself costs 
money. 

H. J. SCHLAFLY: Do you find that there 
are any other advantages of the longi- 
tudinally shielded PSV over the spirally 
shielded PSV? 

MR. MORRISON: It is just a matter of 
the shielding that we obtain. Whereas we 
might get 120 db on one type as a reference 
number, we might get 20 db more on the 
longitudinally shielded conductor. It is 
just a matter of making a good tight cop- 
per shield around the conductor. The 
present manufacture is longitudinally 
seamed plus a winding as you saw in the 
second figure a spiral tape of copper. It 
is shielding to external fields. That is all 
we gain by it. 

E. A. HUNGERFORD, JR. : I was wonder- 
ing if you could tell us anything about the 
possible demand for r-f distribution of tele- 
vision signals to advertising agencies and 
so on in New York City where you could 
get away from going through the air. 
Has the company had a requirement for 
that yet? 

MR. MORRISON: There are always a 
certain number of requirements for experi- 
mental purposes that come in at the front 
or the back door. All though this, people 
are interested in point-to-point connec- 
tions of television. There has been no real 
rush to connect up the city with a form of 
entertainment like Muzak in sound [dis- 
tribution]. However, it is certainly within 
the realm of possibility, I believe; and for 
certain commercial and industrial pur- 
poses, the ability to connect up a video 
facility as you would connect up a tele- 
phone may prove to be rather attractive. 
But so far there is no commercial applica- 
tion of that sort of thing, none at least 
with which I am familiar. 



294 



March 1951 Journal of the SMPTE Vol. 56 



A Professional Magnetic-Recording 

System for Use With 

35> 17V- and 16-Mm Films 

By G. R. Crane, J. G. Frayne and E. W. Templin 



This paper describes a portable magnetic-recording system for producing 
high-quality sound track in synchronism with pictures. The system has 
been designed to enable magnetic recording to conform with standard 
motion picture studio operating practices. A number of features such as 
high-speed rewind, interlocked-switching facilities, one basic type of 
amplifier and the use of miniature tubes throughout have been incorpo- 
rated in the system. 



THE MAGNETIC-RECORDING SYSTEM 
described in this paper is an evolu- 
tionary development of the application 
of magnetic-recording techniques for 
motion picture purposes. In previous 
papers presented before the Society, 1 - 2 
complete descriptions have been given of 
the various supplemental magnetic- 
recording facilities which have been 
made available to the motion picture 
industry by Western Electric Company 
and the Westrex Corporation. The 
widespread use of these modified photo- 
graphic recorders has provided an un- 
usual opportunity to evaluate the per- 
formance under actual field conditions 
of the various circuits and mechanical 
devices incorporated in these modifica- 
tions. In the design, therefore, of this 
completely new magnetic-recording sys- 
tem, those elements have been retained 

Presented on October 18, 1950, at the 
Society's Convention at Lake Placid, 
N.Y., by G. R. Crane, J. G. Frayne and 
E. W. Templin, Westrex Corp., 6601 
Romaine St., Hollywood 38, Calif. 



that have proved to be of definite value, 
while others that have proved super- 
fluous have been eliminated. 

It was stipulated that the new mag- 
netic-recording system should be cap- 
able of being operated with any of the 
known motor systems commonly em- 
ployed in the motion picture industry. 
These include single-phase and three- 
phase synchronous, a-c and d-c inter- 
lock, as well as multiduty operation. 3 
The motor-control circuits of the re- 
corder should be arranged so that in 
order to change from any one motor to 
another, it is only necessary to discon- 
nect and remove the motor, replace it 
with the other type and reconnect it 
with a minimum of effort. 

Before establishing a suitable sound- 
track position for this system, it was 
decided to adhere to what may become 
an industry-wide accepted standard. 
To this end, there has been active co- 
operation with the Motion Picture Re- 
search Council and the Magnetic Sub- 
committee of the SMPTE to try to 



March 1951 Journal of the SMPTE Vol. 56 



295 



TRAVIL 



RCCOft&ING Off. 
f}Cf>OOUCIH6 




Fig. 1. Proposed magnetic film-track standards. 



establish a single- and multiple-track 
standard for 35-, 17 ' l / 2 - and 16-mm 
films. Unfortunately, the proposals 
of both of these organizations have not 
yet reached the stage of complete 
standardization, but the response of 
the great majority of equipment manu- 
facturers and recording studios has been 
so favorable that it was decided to pro- 
ceed on the assumption that they would 
eventually be accepted as standard. 

The American Standards Association 
and the Research Council track pro- 
posal is shown in Fig. 1. It provides 
for three 200-mil tracks, with a separa- 
tion of 150 mils between tracks and a 
50-mil separation between outer tracks 
and sprocket holes. Track No. 1 on 
the left of the figure becomes the normal 
single track, and it will be noted that 
it is the so-called negative sound-track 
position. It differs from the original 
track location previously described in 
that it has been moved in from the 136- 
mil separation from the sprocket-hole 
side to the present 50 mils. This was 
done to permit the later development of 
multitrack recording in accordance 



with Fig. 1. It should be noted that 
although the recorder described below 
is designed to meet the new proposed 
track standard, provisions are incor- 
porated for restoring the older track 
location when desired. 

The situation with regard to 173^-mm 
film track standards is not as clear as with 
35-mm film track. The original proposal 
was to record down the slit edge of this 
film. This would correspond to a posi- 
tive sound-track position for this film. 
A later proposal called for the 17^-mm 
track to be in an identical position with 
that of the 35-mm film. In the re- 
corder described below, provision has 
been made to record the 17^-mm 
track at either of these positions, and 
experience alone can tell which is more 
likely to be adopted on an industry- 
wide basis. 

The track position for 16-mm film 
has been set at 6 mils from the unper- 
forated edge and no request has been 
made so far for any variation from this 
position. In all three cases, the track 
width is 200 mils and the same record- 
ing head is used for all three positions. 



296 



March 1951 Journal of the SMPTE Vol. 56 



COMMON SPRING 



SPROCKET 




SPROCKET 



^INTERNAL FILTER CIRCUIT 
Fig. 2. Film-drive schematic. 



It was decided that provision for 
high-speed film runback should be 
available in this system. This was 
considered a necessity in the foreign 
field and a desirable accessory in the 
domestic field. An actual runback 
speed of 3X has been provided in the 
system described in this paper. As will 
be seen later, this is accomplished with- 
out unthreading the film in the recorder. 

Recorder 

The recording machine has been de- 
signed primarily for portable use as a 
production recorder for magnetic sound 
in synchronism with a motion picture. 
The machine and its control features 
are readily adaptable to any of the cur- 
rent types of motor-drive systems in- 
cluding interlock, multiduty and syn- 



chronous, either 1 </>, 115 v or 3 <, 220 v. 
In addition, it may be readily equipped 
for operation with 35-, 17^- or 16-mm 
film which changes only a few parts of 
the recorder, such as rollers, sprockets 
and reel shafts. 

Recorder Film Path 

The film drive is of the Davis type, 
which was discussed in an earlier paper 
published in the JOURNAL, 4 and con- 
sists of two 16-tooth sprockets with a 
symmetrical, tensioned path between 
them, which includes two filter rollers 
and an impedance drum within which 
is located the record head. The path 
also includes a slight wrap around a 
monitor magnetic head beyond the 
drum, and a roller which may be re- 
placed by an optional erase head just 



Crane, Frayne and Templin: Magnetic Recording on Film 



297 



ahead of the drum. The film-path sche- 
matic and its equivalent electrical cir- 
cuit are shown in Fig. 2. This method 
of mechanical filtering is, of course, 
in effect a low-pass filter and the 
damping is provided by an oil dash- 
pot connected to one arm to give ap- 
proximately critical damping at its 
natural resonant period of about 1.6 sec. 
The film path and its elements are 
essentially the same for 17^- and 16- 
mm film. In the case of the narrower 
film widths, the lateral position of the 
magnetic heads is not changed and the 
roller flanges are placed so that the film 
is properly registered with the heads 
for correct track position. In the case 



of 17^-mm film, the track may be 
recorded either adjacent to the split edge 
of the film or in the Academy proposed- 
standard position. 

Figure 3 shows the front of the ma- 
chine with the cover over the magnetic 
heads removed. Two shock rollers are 
provided over which the film passes 
coming into the feed sprocket and leav- 
ing the holdback sprocket. These 
rollers have relatively low inertia and 
are controlled by a spring and a single 
oil dashpot. They protect the film 
sprocket holes from abuse, particularly 
when the machine is started or when 
there is any abnormal operation such as 
jerks or erratic motion imparted to the 




Fig. 3. Front view of RA-1467 Recorder with cover of head assembly removed. 
298 March 1951 Journal of the SMPTE Vol. 56 



film by imperfect reels. The right- 
hand shock roller serves also as an 
"antibuckle" device or indicator of loss 
of take-up tension. The details of this 
function will be described later. 

The film sprockets have relatively 
large teeth which fit rather closely into 
the sprocket holes and take advantage 
of the low-shrinkage characteristics of 
the new acetate-base film. The tooth 



clearance in the hole is still sufficient 
to permit satisfactory operation with 
moderately shrunk film as high as 0.6%, 
but the clearance is small enough to 
eliminate most of the so-called crossover 
effect caused by the changing film ten- 
sions on the two sides of the sprocket 
causing the sprocket to act alternately 
as a feed and holdback sprocket. Both 
sprockets have flanges as an aid to rapid 




-s,. 







Fig. 4. Rear view of RA-1467 Recorder. 
Crane, Frayne and Templin : Magnetic Recording on Film 



299 



threading and this has permitted the 
use of a simple film shoe instead of the 
usual sprocket pad-roller assembly. 
This shoe is adjusted so that it clears 
the film surface during normal opera- 
tion. It serves only as a guard for 
such conditions as starting and stopping 
the equipment. Each sprocket also 
has a knob by which the machine may 
be turned over when the selector knob 
is in the "Neutral" position. Since the 
motor is disengaged, this feature is 
particularly useful in cases where the 
machine is used as a dubbing reproducer 
and a start mark must be registered 
without disturbing the position of the 
interlock motor. 

The film passes in and out of the ma- 
chine to reels located above, with a 
single, round belt driving both reel 
shafts, which may be seen in Fig. 4. 
Convertible overrunning-clutch assem- 
blies are provided to permit either reel 
to rotate in either direction to meet the 
varying practices with regard to direc- 
tion of rotation of feed and take-up 
reels now prevalent in the industry. 
The same belt may be used with alter- 
nate crossed paths to change the rota- 
tion. 

Recorder Controls 

The motor is controlled by a d-c relay 
of the mechanical-latch type with push 
buttons for start and stop. This sys- 
tem has the double advantage of having 
no power in the relay during operation 
of the machine, thereby simplifying the 
magnetic shielding problem and provid- 
ing a convenient method for controlling 
the recorder remotely with momentary- 
contact switches carrying only relay-coil 
current. 

The machine functions are controlled 
by a single selector knob on the front, 
as shown in Fig. 3. It is mechanically 
linked to the gear box for speed selec- 
tion. It operates a micros witch for 
speech and bias disconnect during play- 
back and rewind, and also disables the 
antibuckle device during rewind. Erase 



facilities are controlled in a similar man- 
ner if used. 

The shock roller which is used for 
indication of take-up failure occupies 
the same position at rest and at take-up 
failure; therefore, the operating circuit 
is not energized until 3 sec after start. 
A time-delay relay across the motor 
circuit performs this function as well as 
operating the relay to transfer the 
recordist's monitor from direct to film 
monitor after 3 sec from start. In 
addition, the delay relay has contacts 
closing after 1 sec to short resistors in 
the motor line to reduce the high ac- 
celeration of certain types of synchro- 
nous motors. All relays are, of course, 
reset instantaneously when power is 
removed. 

In the event of take-up failure, the 
motor is lifted from the line and the 
main transmission circuit is disabled 
which removes the signal from the 
mixer's volume indicator and the direct 
monitor line. Since the recorder is 
stopped, film monitoring is also ter- 
minated. The buckle condition is re- 
stored only by operation of the recorder 
power switch to "Off," but the motor 
is not reconnected to the line until the 
motor "Start" button is operated again. 

To provide the proper correlation of 
the synchronous-motor starting re- 
sistors, the time-delay relay voltage re- 
quirements, and other circuit functions, 
a four-position switch with a screw- 
driver-slot control appears on the rear 
panel of the recorder. This switch is 
set to the indicated position for any one 
of the various types of motor systems, 
thereby making all of the necessary cir- 
cuit changes. 

Recorder Structure 

The upper assembly of the recorder 
containing the two reel shafts is remov- 
able from the recorder case by three 
thumbscrews. The take-up belt is 
pushed back into the recorder and 
covered with a sliding cover. Space is 
provided in the control unit for contain- 



300 



March 1951 Journal of the SMPTK Vol. 56 



ing the reel assembly when the system is 
to be transported. The recorder case 
has a removable rear cover for access to 
the motor-starting resistors and other 
components, and contains a recessed 
opening through which all of the cords 
may be inserted into the recorder re- 
ceptacles. A front cover with a trans- 
parent window is used primarily for a 
dust cover during stand-by or for ship- 
ment. It is also useful in those cases 
where the recorder is to be operated near 
the action, thereby requiring further 
reduction of recorder noise and that 
caused by the film engagement on the 
sprockets. An accessory magazine is 
also available for completely enclosing 
the two film reels for further reduction 
of noise caused by film scuffing on reel 
flanges. This magazine is demountable 
and has transparent doors for visibility. 

Mechanical Drive 

As previously mentioned, various 
types of motors may be accommodated 
and the one selected is directly coupled 
through a torsionally rigid, flexible 
coupling to a gear box. 

Between the gear-box output shaft 
and a cross shaft, interchangeable sets 
of 90 helical-change gears are used to 
accommodate all currently used motor 
speeds from 1000 to 1800 rpm for either 
35-mm or 16-mm film speeds. The 
cross shaft drives each of the two 
sprocket shafts through similar 90 
helical gears. Each of these three sets 
of gears has a nylon plastic gear driven 
from a steel gear which gives quiet and 
smooth operation. Nylon was chosen 
as the nonmetallic material since it has 
unusual properties suitable to this appli- 
cation. It is capable of running with 
virtually no lubrication and performs 
very well over long periods of time with 
a minimum of lubrication provided by 
a drop or two of a special oil which 
clings to the tooth surfaces with high 
tenacity. The material is extremely 
tough and accepts considerable abuse 
in shock loading without damage. 



The gear box accomplishes a 3:1 
speed change by means of planetary 
gears and the ratio change is accom- 
plished by a spring-loaded control rod 
protruding from the center of the 
driven shaft. 

The take-up clutches associated with 
each reel shaft contain overrunning 
clutches as well as a frictional drag to 
the frame of the machine, so that no 
attention need be given to the take-up 
performance regardless of the direction 
or speed of the recorder. The take-up 
clutch provides the proper film take-up 
tension on the reel which requires it, 
and a second small clutch places a light 
drag on the feed-reel shaft to insure 
stable operation. 

The impedance drum, the two 
sprocket-shaft assemblies and the pad- 
roller assemblies have their ball bear- 
ings contained in tubular subassemblies 
so that their lateral position may be 
easily adjusted and locked by means of 
set screws. All rollers have their shafts 
arranged so that they may be likewise 
adjusted laterally. These facilities per- 
mit changes and alignment adjustments 
in film-path components to be readily 
made with a minimum of effort. Until 
track positions are more universally 
standardized and accepted, this feature 
may be useful. 

The head assembly containing the 
two magnetic heads, or three in the case 
of the erase option, may be removed as 
a complete unit and replaced on dowel 
pins without disturbing any of the head 
adjustments relative to the impedance 
drum. The wiring between the mag- 
netic heads and the recorder is termi- 
nated through a small symmetrical 
seven-pin plug arranged to permit a 
180 turnover. This reverses record 
and moniter connections so that the 
monitor head may be used in the rare 
event of failure at the record-head posi- 
tion or the record head may be used for 
high-quality reproduction as in the 
case of transfer machines or high-quality 
playback. The record and monitor 



Crane, Frayne and Templin: Magnetic Recording on Film 



301 



'!- 5.^211. 
u "CONTROLS* 



&3VI 



B c 




FOR 
3d SYNCHRONOUS MOTOR 

INTERLOCK MOTOR 
MULTI-DUTY MOTOR 



C=3 



FOR OPERATION 
FROM D-C 



Cjp 

/K 

OX 
II5V. 



Fig. 5. System block schematic. 



heads have mountings equipped with 
vernier-screw adjustments for setting 
azimuth quickly and accurately, and the 
solidly mounted record head has a 
vernier adjustment for its position rela- 
tive to the impedance drum and the 
film. The head curvature lying within 
the film curvature determined by the 
impedance drum insures excellent con- 
tact with the magnetic coating and its 
position is such as to insure a long 
period of service without requiring any 
readjustment for wear compensations. 

Transmission System 

An over-all block schematic of the 
system is shown in Fig. 5. The three 
basic units are shown: the Recorder, 
the Mixer and the Control Unit. The 
control unit is normally associated 
closely with the recorder and connected 
to it by two 10-ft interconnecting cables. 
The motor cable also connects directly 
between the recorder and control unit 
when a 115-v, 1-0 drive motor is used. 
Only one interconnecting cable is re- 
quired between the mixer and the con- 



trol unit. The separation between 
these units may be 50, 100 or 150 ft 
with no special provisions required for 
normal variations in power-supply volt- 
age and voltage drop in the inter- 
connecting cable. An additional cable 
from recorder to studio facilities may be 
used to provide motor start-stop con- 
trols, speed signal and footage-counter 
control at remote points. 

The transmission system is built up 
of combinations of three basic types of 
electronic subassembly components: an 
amplifier, an oscillator and a power 
supply. This method of building up 
the system from a minimum number of 
standardized types of subassemblies 
has several advantages, including econ- 
omy of manufacture, simplicity of main- 
tenance and a minimum investment in 
studio plant and location spares. 

Amplifier 

Only one type of amplifier is used 
throughout the recording system. A 
total of four are used in the system, one 
each for the two microphone preampli- 



302 



March 1951 Journal of the SMPTE Vol. 56 



fiers, one for the main recording ampli- 
fier and one for film monitor. Only one 
type of vacuum tube is used in all the 
amplifier applications the G.E. 12AY7 
miniature twin triode. The special per- 
formance requirements for the particu- 
lar amplifier applications are all ac- 
commodated by a series of plug-in units 
which make connections to internal cir- 
cuits of the amplifier. 

With suitably filtered power supply 
for plates and heaters and with a reason- 
able amount of selection of tubes for the 
input stage, a noise level of ap- 
proximately 125 dbm, referred to the 
amplifier input, may be obtained. This 
permits a signal-to-noise ratio of ap- 
proximately 55 db for normal dialogue 
pickup from a W.E. RA-1142 Micro- 
phone. 

The amplifier will carry an output 
power level of +22 dbm for 1% dis- 
tortion which provides a comfortable 
margin over that required for both re- 
cording and monitoring applications. 

The power requirements are 10 ma 
(milliamperes) at 275 v d-c and 0.3 amp 
at 12 v. A d-c or rectified a-c heater 
supply is recommended for all low-level 
applications. 

Bias Oscillator 

The oscillator is of the L-C tuned-grid 
type, operating at 60 kc and employing 
one 12AU7 twin triode operating in 
push-pull. The total distortion appear- 
ing at the oscillator output terminals is 
less than Ho of 1%. The oscillator will 
deliver at least 35 ma at 60 kc into the 
record head. The power requirements 
are 6 ma at 275 v d-c and 0.15 amp at 
12 v a-c or d-c. 

Power Supply 

The power supply provides line-and- 
load-regulated plate current and unregu- 
lated heater current for the entire 
system. It requires 1 amp from a 50- 
or 60-cycle 115-v power source. 

The circuit includes a 3-stage d-c 
amplifier and a series regulating tube. 
Regulation over a line-voltage range of 



10% and a load range of to 55 ma 
is obtained with a maximum of not more 
than 0.5 v variation in output. The 
total power-supply ripple is approxi- 
mately 0.5 mv or less over the complete 
load range. The a-c impedance of the 
output is also held to a very low value, 
thus simplifying the decoupling require- 
ments between stages of an individual 
amplifier as well as between high- and 
low-level amplifiers. The output volt- 
age is adjustable over a range of 255 to 
300 v but is normally intended to be set 
to 275 v. 

A bridge-type selenium-cell rectifier 
is used to provide 12 v d-c for vacuum- 
tube heater and relay control circuits. 
This supply is biased 20 v above ground 
which makes the vacuum tubes in low- 
level stages less sensitive to residual 
power-frequency ripple and simplifies the 
filtering requirements . This supply pro- 
vides 1.8 amp at 12 v with a ripple less 
than 1 v. For a ="=10% variation in 
power-supply voltage and for mixer 
cable lengths up to more than 100 ft, 
the voltage at the heaters is within safe 
operating limits without special regu- 
lating or current-limiting provisions. 

Mixer 

The mixer is a complete speech-input 
equipment having two microphone in- 
puts and supplying signal directly to 
the recording head, direct-monitor lines 
and volume indicator. Figure 6 is a 
view of the mixer. 

Three of the basic amplifiers previ- 
ously described are used in the mixer, 
two as microphone preamplifiers and 
one as the recording amplifier. 

For the preamplifier application, the 
plug-in unit inserts variable dialogue 
equalization and low-frequency pre- 
equalization. The dialogue-equalizer 
characteristic is selected by a control 
knob on the top of the plug-in unit. 
The response curves are shown in Fig. 7. 
One position provides the normal flat 
amplifier characteristic; the other two 
provide, respectively, 8- or 12-db droop 



Crane, Frayne and Templin: Magnetic Recording on Film 



303 



at 100 c(cyles per sec). These char- 
acteristics follow the conventional ones 
that have been used for many years in 
Hollywood studios. Below the useful 
dialogue range they maintain sufficient 
loss so that a high-pass filter is not 
normally required. This is particularly 
the case since the low-frequency diffi- 
culties in photographic recording at- 
tributable to noise-reduction and peak- 
limiting operations are inherently ab- 
sent. 

The low-frequency pre- and post- 
equalization used in the system takes 
advantage of the energy-distribution 
characteristic of speech and music 5 
to increase the margin between signal 
and residual-hum components in the 
reproducing or monitor system. As 



shown in Fig. 7, the pre-equalization 
amounts to a 2^-db rise at 50 c. 

The design of the equalizers is such 
that the gain in the region of 1000 c is 
essentially unchanged for all settings of 
the dialogue equalizer and with the low- 
frequency pre-equalizer in or out of 
circuit. 

The plug-in unit also contains re- 
sistive elements which introduce attenu- 
ation in the amplifier circuits terminat- 
ing therein. The gain of each stage is 
thereby carefully established at the 
value giving the best possible balance 
between signal-to-noise and margin- 
from-overload, based on the sensitivity 
of the microphone and the range of in- 
put level to be accommodated. The 
midfrequency gain of the amplifier as 




304 



Fig. 6. View of KA-1485-A Mixer. 
March 1951 Journal of the SMPTE Vol. 56 



6 



S-io 












Odb.STEP^l 

^-SdbSTEP > WITHOUT PRE-EQUALIZATION 
H2dbSTEpJ 



RA-I474A AMPLIFIER WITH ASO-76213 P. I.UNIT 
DIALOGUE & LOW FREQUENCY PRE-EQUALIZATI 



20 



50 



100 



200 500 1000 2000 5000 10000 20000 

FREQUENCY IN CYCLES PER SECOND 

Fig. 7. Preamplifier gain-frequency characteristics. 



-HO 




^A-I474A AMPLIFIER WITH ASO-76214 P. I. UNIT 
MID-RANGE EQUALIZATION 



20 



50 



100 



200 500 1000 200O 5000 10000 20000 

FREQUENCY IN CYCLES PER SECOND 
Fig. 8. Recording-amplifier gain-frequency characteristics. 



established by the plug-in unit is 53 db. 

The preamplifiers are followed by 
separate microphone cutoff keys, mixer 
pots, a combining network and a gain 
switch having three 10-db steps. 

Following this is the recording ampli- 
fier with its input connected for un- 
balanced, 600-ohm, terminated opera- 



tion. The plug-in unit in this applica- 
tion provides two steps of midrange 
equalization in addition to the flat 
characteristic, a choice of the three con- 
ditions being selected by the control 
knob in the top. As shown in Fig. 8, 
this equalization consists of a rather 
broad peak of either 4 or 7 db centered 



Crane, Frayne and Templin: Magnetic Recording on Film 



305 



near 5000 c. This boost is used on 
dialogue only, supplementing the rise 
in this same region introduced by many 
of the regularly used microphones. Its 
result is to improve the "presence" 
quality of the reproduced speech at the 
listening levels encountered under 
theater reproducing conditions. The 
gain of this amplifier is established in 
the plug-in unit at 58 db, and for fre- 
quencies below 1000 c it is maintained 
at this value for all positions of the mid- 
range-equalization control switch. 

The 600-ohm output is partially 
loaded by a 1000-ohm resistor (located 
in the control unit) which feeds the re- 
cording head. The 1000 ohms is large 
compared with the impedance of the 
head and causes the current in the head 
to be substantially independent of the 
head impedance throughout the audible 
frequency range. The volume indi- 
cator is also across this 600-ohm output. 
The remainder of the amplifier loading 
is on the 50-ohm output supplying direct 
monitor for the mixer. 

Direct and film monitor are available 
in the mixer. However, due to the 
fractional-second time delay in the film 
monitor, it is expected the mixer 
operator will normally listen on the 
direct line, with only occasional check- 
ing from the recorded film. 

The volume indicator has a high- 
speed movement and new design pro- 
viding increased sensitivity and less 
bridging loss than those previously used. 
Its maximum sensitivity is dbm for 
db meter deflection and its internal im- 
pedance is such that it may be used at 
this setting under operating conditions. 

Control Unit 

The control unit contains miscellane- 
ous components including the bias 
oscillator, film-monitor amplifier, power 
supply and interconnecting circuits be- 
tween the mixer and the recorder. It 
also provides storage space for the film- 
reel assembly which mounts on the 
recorder during operations. The out- 



put from the mixer is carried directly to 
the recorder for recordist direct monitor 
and through the 1000-ohm resistor and a 
60-kc suppressor filter to the record 
head. The 60-kc filter prevents the 
bias signal from affecting the volume 
indicator. In the event of a film buckle 
or take-up failure, this direct-monitor 
line is shorted by the antibuckle relay 
described earlier. This also shorts the 
signal to the volume indicator and 
recordist direct monitor and thus serves 
as a warning to the mixer operator. 

The plug-in unit for this application 
contains a continuously adjustable gain 
control, for balancing film and direct- 
monitor levels, and a reproducing 
equalizer. The latter has the con- 
ventional 6-db-per-octave slope plus 
low-frequency postequalization comple- 
mentary to the pre-equalizer, and high- 
frequency equalization complementary 
to the magnetic losses inherent in the 
recording-reproducing process. 

Space is available for substituting a 
bias-erase oscillator operating directly 
from the 115-v power source when the 
erasing facility at the recorder is re- 
quired. 

The size of the control unit was 
chosen to permit its use as a mounting 
support for the recorder during opera- 
tions. 

System Performance 

The 100%-modulation recording level, 
as measured at the volume indicator in 
the mixer unit, is determined by making 
a series of measurements of output level 
and per cent distortion at the repro- 
ducing-amplifier output for various 
values of recording level and bias cur- 
rent. A typical set of data for a particu- 
lar recording head and film emulsion is 
shown in Fig. 9. Based on an allow- 
able total harmonic distortion of 3%, it 
can be deduced from the curves that a 
100% modulation recording level of 
==10 dbm, with a bias current of ap- 
proximately 25 ma, is optimum. Lower 



306 



March 1951 Journal of the SMPTE Vol. 56 



^ o 



-2.0 



-4.0 



- 6.0 



-8.0 



-12.0 



V.I. +12 



V.I. +10 
V.I. +8 



V.I. + 6 



BIAS VS. OUTPUT 



10 12 14 16 18 20 22 24 26 28 30 32 34 



RECORDED 
LEVELS 



BIAS CURRENT (MA) 



BIAS VS. DISTORTION 




8 A 10 

RECORDED 
LEVELS 



12 14 



16 18 20 22 24 26 28 30 32 34 
BIAS CURRENT (MA) 



Fig. 9. Magnetic-recording characteristics. 



recording levels give a lower reproducing 
level with corresponding decrease in 
signal-to-noise ratio. Higher recording 
levels require increased bias current for 
the permissible amount of distortion, 
with no appreciable increase in output 
level. For this optimum operating 
condition, a level of 4 dbm is available 
at the recorder on either direct or film 
monitor. 

By means of the mixer pots and main 
gain control, a 60-db range of input level 



may be held to the 100% modulation 
level of the system. For any combina- 
tion of mixer and gain settings, the 
carrying capacity will be limited by the 
film medium rather than by the trans- 
mission equipment. 

The signal-to-noise ratio of the re- 
cording circuit as limited by the first 
stage of the preamplifier is approxi- 
mately 55 db for normal dialogue. 

The over-all record-reproduce film 
characteristic for 35- or 17^-mm films, 



Crane, Frayne and Templin: Magnetic Recording on Film 



307 



RELATIVE CURRENT IN RECORDING HEAD 



RELATIVE RESPONSE FROM 




00 200 500 1000 2000 

FREQUENCY IN CYCLES PER SECOND 
Fig. 10. 16-mm frequency characteristics. 



5000 10000 



for flat input to the mixer and, exclud- 
ing the dialogue and midrange equaliza- 
tion, is essentially flat from 50 to 7000 c, 
which is more than ample for monitor- 
ing purposes. For high-quality re- 
recording, the flat response may be ex- 
tended upward to 10,000 c by using the 
film-loss equalizers, normally a part of 
existing photographic-magnetic re-re- 
corders. A signal-to-noise ratio from 
the reproduced film of approximately 
55 db may be obtained. For special 
applications, this may be increased to 
60 db or more as has been done in earlier 
photographic-magnetic equipment 1 by 
additional low- and high-frequency pre- 
and post-equalization. 

Where high-frequency pre-equaliza- 
tion is to be utilized for this further in- 
crease in signal-to-noise ratio, an ap- 
preciable portion of it can be considered 
as precompensation for the magnetic 
and scanning losses inherent in the re- 
cording-reproducing process. These 
latter losses then introduce the com- 
pensating post-equalization to provide 
the flat over-all record-reproduce char- 



acteristic. Thus, if the high-frequency 
pre-equalization is held to that value 
required to precompensate for these 
losses, an electrical high-frequency post- 
equalizer is not required and flat re- 
sponse up to approximately 9000 c may 
be obtained. For 35-mm or 17^- 
mm film, this pre-emphasis has been 
obtained by a condenser-resistance com- 
bination shunted across the 1000-ohm 
resistance in the recording-head circuit. 
For 16-mm film, with its inherently 
lower cutoff frequency, a series-tuned 
circuit is bridged across the resistor in 
series with the head to provide a high- 
frequency pre-equalization character- 
istic as shown in Fig. 10. A typical 
gain-frequency response from film re- 
corded with this characteristic and re- 
produced on a high-quality re-recorder 
is also shown in Fig. 10. The response 
is substantially flat to approximately 
6500 c, except for the low end which 
contains reproducing post-equalization 
to compensate for the low-end pre- 
equalization generally used in Western 
Electric magnetic recording systems. 1 



308 



March 1951 Journal of the SMPTE Vol. 56 



References 

1. Lewis B. Browder, "Direct-positive 
variable-area recording with the light 
valve," Jour. SMPE, vol. 53, pp. 149- 
158, Aug. 1949. 

2. G. R. Crane, J. G. Frayne and E. W. 
Templin, "Supplementary magnetic 
facilities for photographic sound sys- 
tems," Jour. SMPTE, vol. 54, pp. 
315-327, Mar. 1950. 

3. A. L. Holcomb, "Motor systems for 
motion picture production," Jour. 
SMPE, vol. 42, pp. 9-33, Jan. 1944. 

4. C. C. Davis, "An improved film-drive 
filter mechanism," Jour. SMPE, vol. 
46, pp. 454-464, June 1946. 

5. L. J. Sivian, H. K. Dunn and S. D. 
White, "Absolute amplitudes and spec- 
tra of certain musical instruments and 
orchestras," /. Acous. Soc. Amer., 
vol. 2, p. 330, Jan. 1931. 

Discussion 

E. W. KELLOGG: With regard to the 
Davis Drive mentioned in the paper, 1 
don't know just what it is intended to 
cover by that name, or what features are 
to be credited to Mr. Davis. I wish to 
call attention to the fact that if, broadly, 
the filter system means a solid flywheel 
on the shaft of a drum, and compliance 
introduced between the driving sprockets 
and the drum by means of a flexibly 
mounted idler-roller, that system dates 
back to ancient history. It was shown in 
a Triergon patent filed in the United 
States in 1922. It was again shown in 
slightly more definite form in a patent to 
Poulsen and Petersen, under which an 
unsuccessful infringement suit was brought 
against RCA Mfg. Co. and Electrical Re- 
search Products, in the late 1930's. It is 
a very effective filter, and I think all 
fundamental patents on it have long since 
run out. The feature of connecting a 
dash-pot to the movable idler is described 
in a 1931 paper of mine, describing the 
first RCA magnetic-drive recorder and 
also in the corresponding patent, in which 
broad claims were allowed on damping. 
I think that we should, in attaching 
anyone's name to a filter system, make it 
clear that it does not comprise those 
broad features that I have just described. 
Perhaps you can enlighten us about the 
features that have been added in the way 
of refinements or improvements. 



In the machine just described, what is 
the position of the recording or reproduc- 
ing magnet in relation to the drum? 

DR. FRAYNE: In connection with Dr. 
Kellogg's question, all the data on the 
track width and location will be found in 
the published paper. I would say this, 
that the recording part of the magnetic 
head is placed as close to the drum as is 
mechanically feasible. Under those con- 
ditions we have no trouble whatever with 
quick starting of the drum or other prob- 
lems relating to velocity or amplitude 
modulation. The flutter in this machine, 
by the as yet nonaccepted standard, 
measures somewhat less than % of 1 per 
cent and the flutter rates are practically 
all above 100 cycles, as is customarily 
found in most magnetic recorders. With 
regard to the Davis Drive, we are quite 
cognizant of the contributions of Dr. 
Kellogg and others. It is to be regretted, 
however, that they never saw fit to intro- 
duce it to the industry. The Western 
Electric Company, as far as I know, was 
the first to introduce the tight-loop type 
of drive and the Davis Drive is so called 
for the reason that it was recognized by 
the Academy and given an award; and in 
that award, the name "Davis Drive" was 
created. The name "Davis" was not 
given to the drive by the Western Electric 
Company. The Davis Drive itself is 
covered by a U.S. patent, the principal 
patentable feature being the common 
spring connecting the two compliant 
rollers. The tight loop thus created 
eliminates the necessity of the customary 
pressure-pad roller. Shortly after we 
introduced this tight-loop drive, our 
competitors brought out a similar one. 
Perhaps Dr. Kellogg could enlighten us 
on why that happened. 

DR. KELLOGG: Dr. Frayne has brought 
up the matter of the introduction to the 
industry of the filtering system which 
depends on damped movable idler-rollers. 
Certain obstacles to the use of this system 
were cleared away in the early 1940' s, 
but the war years were not the time to 
jump to new models, particularly when 
the older ones were doing well, which was 
certainly true of the RCA rotary sta- 
bilizer soundheads. But our first postwar 
recorders, and 16-mm projectors, utilized 
filters of the damped, movable-idler type. 



Crane, Frayne and Templin: Magnetic Recording on Film 



309 



Carbon Arc Characteristics That 
Determine Motion Picture Screen Light 

By M. T. Jones and F. T. Bowditch 



In a carbon arc motion picture projector, definite relations exist between 
screen light on the one hand, and the arc current, current density, carbon 
size and the speed and collection angle of the projector optical system on the 
other. Measurements on more than 100 standard and experimental 
carbon arcs, with carbons ranging in size from 9 mm to 16 mm, have pro- 
vided data to establish these relationships. Conditions are denned which 
are of importance in the matching of an optical system to a given arc, or 
vice versa, and for obtaining optimum performance in any situation 
involving screen distribution, amount of light and preferred current. 



IN AN EARLIER PAPER 1 a method is de- 
scribed for calculating motion pic- 
ture screen light from measurements of 
brightness over the carbon arc crater as 
viewed from selected angles, and from a 
consideration of the characteristics of 
the particular optical system involved. 
This method has now been applied to a 
variety of standard and experimental 
carbons, and the resulting data analyzed 
to establish certain significant relation- 
ships which form the subject of this 
paper. These relationships are con- 
cerned with the distribution and the 
amount of light delivered to the motion 
picture screen, as these are determined 
by the arc current, the current density, 
the size of carbon and the collection 
angle and speed of the optical system. 
As an illustration of the basic data 

Presented on October 20, 1950, at the 
Society's Convention at Lake Placid, N.Y. 
by M. T. Jones and F. T. Bowditch, 
National Carbon Research Laboratories, 
Box 6087, Cleveland 1, Ohio. 



from which these trends are established, 
calculations made from measurements 
on three experimental trims, each at its 
maximum operating current, are shown 
in Figs. 1, 2 and 3. In this, and in all 
subsequent cases throughout this paper, 
these calculations are made according to 
the method previously described, 1 for 
the one best-focus condition giving 
maximum screen light. Each of these 
curves shows, on the left, the lumens 
through the motion picture aperture 
and, on the right, the light distribution 
across the aperture, each over a range of 
light-collecting angles from the source, 
and for a series of optical speeds into the 
aperture. Light losses due to absorp- 
tion, shadowing and vignetting, which 
always occur in varying degree in any 
specific optical system, have not been 
included in these present calculations, a 
permissible simplification since only rel- 
ative values are considered in the con- 
clusions drawn here. A suitable loss 
correction of approximately 50% would 



310 



March 1951 Journal of the SMPTE Vol. 56 



9MM. CARBON 

AT 180 AMPERES 

(EXPERIMENTAL) 



100% 




60 



80 100 

MIRROR 



120 140 
COLLECTING 



ANGLE 



80 100 120 
IN DEGREES 



160 



Fig. 1. Screen-light characteristics of an experimental 9-mm high-intensity 
positive carbon at its maximum operating current in water-cooled jaws. 

NOTE: All light and distribution values throughout this paper are based upon the best- 
focus condition giving maximum screen light. 



have to be applied to the lumen values 
given in this paper in order to determine 
the actual screen-light level in any par- 
ticular instance. As an example, crater 
light measurements on an 8-mm to 7- 
mm "Suprex" trim at 70 amp, calcu- 
lated for an //2.0 mirror, predict a flux 
of 27,600 1m on the aperture, compared 
with 14,000 1m motion picture screen 
light realized in practice. This is be- 
cause mirror absorption and reflectance 
losses, plus shadowing due to the posi- 
tive head, etc., amount to about 20%; 
while of the total lumens passing the film 
aperture, no more than about 65% 
reaches the screen due to a combination 
of spill-over, vignetting and glass trans- 
mittance losses at the projection lens. 

With respect to the aperture-lumen 
variations shown by Figs. 1, 2 and 3, 
these confirm the earlier conclusion 1 that 
maximum luminous flux is not neces- 
sarily obtained at the maximum collec- 



tion angle; the simple concept that a 
bigger collection angle picks up more 
light from the source and hence delivers 
more light to the motion picture screen 
fails to work out. With a fixed speed 
into the aperture, the optical geometry 
is such that the magnification of the 
crater image on the aperture increases as 
the pickup angle increases, thus intro- 
ducing a loss factor, working against the 
greater light collection. The light dis- 
tribution characteristics of high-inten- 
sity carbon arcs are such that a collec- 
tion angle is reached at each speed be- 
yond which more light is thrown outside 
the aperture by the enlarged image than 
can be collected by the higher pickup 
angle. The exact pickup angle at which 
this maximum light value occurs will de- 
pend in each instance on the particular 
light distribution characteristics of the 
carbon in question. A small carbon, for 
instance, with a peaked light distribu- 



Jones and Bowditch: Carbon-Arc Screen Light 



311 



150 



100 



60 



60 



1 6 MM. CARBON 
WITH SMALL CORE 
AT 210 AMPERES 



100% 




140 160 



80 100 120 140 80 DO 120 

MIRROR COLLECTING ANGLE IN DEGREES 

Fig. 2. Screen-light characteristics of an experimental 16-mm 

high-intensity positive carbon, with small core, at its 

maximum operating current in water-cooled jaws. 



150 



100 



50 






'60 



F/ 



1.3 



16 MM. CARBON 
WITH LARGE CORE 
AT 460 AMPERES 

(EXPERIMENTAL) 




80 100 120 140 80 100 120 

MIRROR COLLECTING ANGLE IN DEGREES 



140 



100% 



Fig. 3. Screen -light characteristics of an experimental 16-mm 

high-intensity positive carbon, with large core, at its 

maximum operating current in water-cooled jaws. 



312 



March 1951 Journal of the SMPTE Vol. 56 



tion, effectively utilizes a higher magni- 
fication ratio and hence a higher pickup 
angle than is required with a larger car- 
bon with a more uniform light distri- 
bution. 

Figure 1 gives the light characteris- 
tics of an experimental 9-mm carbon op- 
erated at 180 amp, a very high current 
for this size. It is seen that high collec- 
tion angles are effectively utilized at the 
various optical speeds to give good 
screen light values, but at comparatively 
low distribution ratios. Figure 2 shows 
the similar characteristics for an experi- 
mental 16-mm carbon with a small core, 
operated at 210 amp. Here a much 
smaller collection angle gives maximum 
screen light, and the distribution ratios 
are considerably higher. 

Figure 3 shows the light-output char- 
acteristics of another experimental 16- 
mm carbon with a large core, operated 
at 460 amp, the maximum current used 
with any of the approximately 100 posi- 
tive carbons upon which the conclusions 
of this paper are based. Particularly 
with this carbon, the light output and 
distribution ratio are comparatively in- 
sensitive to the choice of collecting 
angle, since, with the large core and 
high current, the effective source is 
quite large and of more uniform bright- 
ness. 

It might be noted that in no case is a 
100% distribution ratio reached. Par- 
ticularly with the large-cored 16-mm 
carbon at //2.0, the effective source size 
is quite sufficient to fill the aperture com- 
pletely from all angles of view. How- 
ever, the crater of any high-intensity 
carbon is always brightest near the cen- 
ter, and this peak is carried through as 
higher illumination in the center of the 
screen. 

Data such as those shown in the pre- 
ceding figures have been correlated for 
approximately 100 different positive 
carbons, both production and experi- 
mental types, of 9-, 11-, 13.6- and 16- 
mm diameter. It is, of course, recog- 
nized that the smaller 7- and 8-mm 



too 




100 200 300 400 
ARC AMPERES 

Fig. 4. Relation between screen 
light and arc current at an optical 
speed of //2.O. 

"Suprex" carbons are very important 
items, commercially, although they 
were not within the scope of the investi- 
gation reported here. Certain basic be- 
haviors have been disclosed by these 
correlations. The first such relation- 
ship is that between screen lumens and 
arc current for various carbon sizes 
and optical speeds. Figure 4 shows this 
relationship at a speed of //2.0 and for 
carbons of 9-, 11-, 13.6- and 16-mm 
diameter. Each curve results from 
measurements on a number of different- 
type carbons of a given size, each carbon 
represented by a single value determined 
at the maximum stable current for that 
carbon. For example, referring to the 
extreme points on the curve for the 
16-mm size, one type of 16-mm positive 
carbon was found to give 32,000 1m at 
its maximum current of 150 amp; 
while another 16-mm positive carbon of 
very different construction gives 68,000 
1m at its maximum current of 460 amp. 
The curves of Fig. 4 show the smallest 
carbon most efficient in current utiliza- 
tion, although, as will be indicated later, 



Jones and Bowditch: Carbon- Arc Screen Light 



313 




800 



600 



400 



200 



9 10 I I 12 13 14 
CARBON DIAMETER IN MM. 

Fig. 5. Current efficiency in 
screen-light production. 

factors other than maximum current 
efficiency are involved in the choice of a 
preferred trim for a particular situation. 
It will be noted that the curves for the 
13.6- and 16-mm carbon sizes are con- 
cave downward, indicating a falling-off 
in current efficiency with increasing am- 
perage on a given size, which is probably 
the result of the inability to cool the 
larger diameters as effectively as the 
smaller. For instance, the 16-mm car- 
bon at 150 amp gives more than 200 
Im/amp; while the larger-cored 16-mm 
carbon at 460 amp gives only 150 1m/ 
amp. This relationship is shown more 
directly by Fig. 5 which utilizes the data 
shown on Fig. 4, together with similar 
data calculated for the other optical 
speeds indicated. Here lumens-per- 
ampere are plotted against carbon di- 
ameter for each of four different optical 
speeds. Each curve is represented as a 
band, including the extremes in current 
efficiency encountered with each carbon 
size. Here again, the higher current 
efficiency of the small-diameter carbon 
is confirmed for each of the optical 
speeds investigated. 

The data so far have been concerned 
only with current efficiency, and if this 
were the only criterion, the smallest 
possible carbon would always be chosen 



100 



80 



60 



40 



20 




2.0/ 



1.6. 



9 10 II 12 13 14 15 16 
CARBON DIAMETER IN MM. 

Fig. 6. Screen-light uniformity; 
side-to-center brightness ratio. 

for a given job. However, no consider- 
ation has yet been given to the screen- 
light distribution ratio, the burning rate 
of the carbon or the color uniformity of 
the screen, all important factors in mak- 
ing a choice in any particular situation. 

Figure 6 shows the variation in screen- 
light distribution ratio with carbon size, 
at the same optical speeds previously 
considered. Here the decided improve- 
ment in screen-light uniformity with in- 
creasing size is effectively demonstrated, 
particularly as the optical speed in- 
creases to give a steeper slope to the 
curve. The data shown in Fig. 6 repre- 
sent the average for all the carbons 
tested, individual values showing some 
scattering around these curves, but not 
sufficient to invalidate the general trend. 
It should be pointed out, however, that, 
contrary to the general indication of 
Fig. 6. all 9-mm carbons, for instance, 
do not yield a lower screen-light distri- 
bution than all of 10-mm size. In fact, 
the reverse is sometimes the case in 
practical service comparisons. Differ- 
ent ratios of core to shell diameter, 
different methods of construction and 
burning, all contribute to the scattering 
previously described. 



314 



March 1951 Journal of the SMPTE Vol. 56 



Two additional factors contribute to 
the screen distribution value actually 
achieved in a given commercial situa- 
tion. The first is due to the slight de- 
parture in shape of all commercial lamp 
mirrors from the perfect ellipse assumed 
in the present calculations. Instead of 
all the crater images from all angles of 
view being precisely centered in the 
aperture, they are displaced in practice, 
by normal errors in mirror shape, to 
spread the light in less peaked fashion, 
but with negligible loss in total lumens 
on account of this spreading. In the 
second place, the projectionist, in ad- 
justing his optics to give the best- 
looking screen, may decide upon a 
slightly out-of-focus setting, and sacri- 
fice somewhat on screen light in favor of 
a flatter screen. The distribution values 
of Fig. 6, therefore, are not necessarily 
the same as those which would be ob- 
tained in a practical projector assembly, 
although the basic trends between sizes 
and optical speeds are as indicated. 

Let us consider next the consumption 
rate of the carbon. This depends so 
much on carbon design, on the method 
of burning, whether the carbon is plated 
or unplated, whether it is burned with 
or without current jaws, and with or 
without water-cooling, that no simple 
relationship exists. However, in situa- 
tions where equivalent screen light is 
given by carbons of different sizes the 
smaller carbon will always burn the 
faster. The exact magnitude and eco- 
nomic significance of this difference re- 
quires determination in each specific 
case, and is always an important factor 
to be considered. Blowing of the arc 
according to principles recently defined 
by Dr. Edgar Gretener, 2 is also a major 
factor in the determination of current 
and carbon efficiency. Apparently the 
light secured at a given current is very 
substantially increased by this blowing, 
while the carbon consumption per unit 
of light output is less markedly affected. 

With respect to screen color, it is 
most difficult to express color differences 



140 



100 



2.0 



\ 



\ 



9 10 I I 12 13 14 15 
DIAMETER OF CARBON IN MM. 

Fig. 7. Collecting angle giving 
maximum screen light. 



in terms of numbers of true comparative 
significance, and no attempt has been 
made to do this with the various trends 
reported here. However, the larger car- 
bon gives a more complete filling of the 
aperture from all angles of view, and 
also tends to give a more uniform screen 
color in any comparison of different 
sizes at equivalent light levels. Further, 
with the larger-sized carbon, screen light 
and color uniformity is better main- 
tained over a wider range of maladjust- 
ment of the positive-carbon position. 

It was previously indicated that the 
smaller carbon requires a higher collec- 
tion angle for maximum screen light 
than does the larger carbon. This gen- 
eral relationship is indicated for four 
different optical speeds by the curves of 
Fig. 7. The increasing slope at the 
higher speed shows that this effect of 
carbon size becomes more pronounced as 
the speed increases. 

Finally, the relationships plotted in 
Fig. 8 show that increases in optical 
speed into the aperture do not result in 
as great increases in illumination as the 



Jones and Bowditch: Carbon-Arc Screen Light 



315 




1.6 SPEED F/1.3 

Fig. 8. Actual versus theoretical 
gain in screen light with increasing 
optical speed. 

relative optical speeds alone would pre- 
dict. Compared to the illumination ob- 
tained with an //2.5 system, an increase 
to //2.0 should theoretically give 6.25/ 
4.00 or 1.56 times as much illumination. 
The ratio calculated with 16-mm car- 
bons is 1.48, and for 9-mm carbons, 1.40 
95% and 90%, respectively, of the 
theoretical amount. As might be ex- 
pected, this departure from the theo- 
retical is greatest for the smallest carbon, 
the reason being that the crater images 
on the aperture are not sufficiently large 
to fill the aperture completely at all 
angles of view, and that the brightness 
distribution across the crater is most 
peaked for the smaller carbons. 



This paper thus defines certain basic 
relationships which should be recog- 
nized in the most effective development 
of the combined arc carbon and optical 
system to do a given job. Broadly 
speaking, a small carbon can be utilized 
to give highest current efficiency; this 
requires the use of a high collection 
angle, gives a less uniform screen-light 
distribution and screen color, and is 
more sensitive to light and color varia- 
tions as the carbon is moved from the 
exact focal position. The larger car- 
bons operate with lower current effi- 
ciency but give a higher quality per- 
formance in all other respects, at a 
higher cost. The choice in a particular 
situation should be based upon a bal- 
ance of these various factors as applied 
to the specific economic considerations 
involved. As in other fields, there are 
proper applications for many possible 
combinations of cost and quality. 

References 

1. M. T. Jones, "Motion picture screen 
light as a function of carbon arc crater 
brightness distribution," Jour. SMPE, 
vol. 49, pp. 218-240, Sept. 1947. 

2. E. Gretener, "Physical principles, de- 
sign and performance of the ventarc 
high-intensity projection lamps," Jour. 
SMPTE, vol. 55, pp. 391-413, Oct. 
1950. 



316 



March 1951 Journal of the SMPTE Vol. 56 



The RCA PT-100 Theater Television 
Equipment 

By Ralph V. Little, Jr. 



The design of the first commercial theater television equipment is based 
on the experience gained from installing and operating earlier develop- 
mental equipments in theaters. Designed to augment the standard 
theater sound equipment, the PT-100 Television Equipment is intended 
to combine maximum reliability with performance limited only by the 
quality of the incoming signal. 



SINCE THE EARLY STAGES of its de- 
velopment, engineers have visual- 
ized television as a natural entertain- 
ment medium for the theater, com- 
parable to that of motion pictures. 
Years of engineering research and de- 
velopment, with special attention di- 
rected to the production of theater-size 
pictures have been rewarded. Theater 
projection television is no longer an en- 
gineer's dream but a current reality. 

The first theater television equip- 
ment was demonstrated in 1929 at the 
RKO Proctor Theater in New York; 
the equipment used a mechanical scan- 
ning disc to produce a 48-line picture 
which was formed by video modulation 
of the light beam from an arc lamp 
source. During the interval of 20 years 
the all-electronic television system was 
devised with the substitution of the 
kinescope for the scanning disc followed 



Presented on October 20, 1950, at the 
Society's Convention at Lake Placid, 
N.Y., by Ralph V. Little, Jr., Radio Cor- 
poration of America, RCA Victor Div., 
Camden, N.J. 



by the addition of the iconoscope for use 
in the camera. 

Paralleling the development of the 
basic technology of electronics, Dr. Ep- 
stein and Mr. Maloff were concentrating 
their efforts on the design of high-in- 
tensity kinescopes and of more effective 
optical systems which led to rapid ad- 
vances in projection television during 
the years 1937-1940. By 1940, a 
modern prototype projection unit 
emerged from the laboratories to be 
demonstrated at the New Yorker 
Theater in 1941. The goal was now in 
view, but World War II delayed further 
development. 

The interest of the film industry was 
enlisted in 1945 and 20th Century-Fox 
and Warner Bros, cooperated to have 
two of the 42-in. optical giants built by 
RCA. These equipments have been de- 
scribed and were demonstrated on 
several occasions during 1947 and one is 
still in use as a standard of excellence at 
the 20th Century-Fox Television Lab- 
oratory. 

With this background of accumulated 
experience in the fields of television and 



March 1951 Journal of the SMPTE Vol. 56 



317 



tube techniques, we were ready to 
establish a product design of a theater 
television system. 

Surveys were made of a number of 
representative theaters to determine the 
physical parameters into which a pro- 
jection system might be integrated. 
There were many factors to be con- 
sidered such as the requirements of per- 
formance, the limitation of present 
theater structures and the economics of 
the purchase and use. 

Using initial prototype designs as a 
basis, a specification was submitted by 
20th Century-Fox as a suggestion of the 
type of equipment which might be 
suited to the ultimate commercial use in 
the theater. Earl Sponable and H. J. 
Schlafly have been actively cooperating 
for four years in the development phases 
of the project and their efforts have 
served in a large measure to bring our 
new design to the industry. Warner 
Bros., through Col. Levenson, con- 
tributed their suggestions. 

The goal of both the engineer and the 
theater industry is to develop theater 
television projection to equal or excel the 
industry standards for 35-mm motion 
picture film projection; under con- 
trolled conditions of pickup and trans- 
mission, the goal appears as a pos- 
sibility. Evolution of an 8-mc video 
channel, used under ideal conditions of 
equipment adjustment, including cor- 
rection for proper tone scale rendition, 
and having a high signal-to-noise ratio, 
should reproduce all the information in a 
frame of 35-mm motion picture film. 
The factors which require attention to 
produce the ultimate in picture accepta- 
bility are: (1) picture detail; (2) free- 



dom from granular or other structure; 
(3) signal-to-noise ratio; and (4) tone- 
scale rendition. 

Picture detail and structure must be 
discussed together as they have to do 
with the transmission bandwidth avail- 
able. The present television broadcast 
channels limit the practical band with, 
via air transmission, to approximately 
4.25 me. The number of scanning lines 
is now 525, permitting a resolution of 
340 lines horizontally and 400 lines 
vertically as seen by use of the Mono- 
scope test pattern. At a 4:1 viewing 
distance, which is the minimum for 
home viewing, the scanning lines cannot 
be resolved by the eye so that the 
standards are considered adequate. 

In order to accommodate the theater 
patrons, who will be closer to the screen 
than the minimum 4:1 distance, more 
picture detail* and a greater number of 
scanning lines will be desirable. The 
selection of the number of scanning lines 
is a function of the economical band- 
width and a compromise on the balance 
between the resolution of the picture 
elements in the horizontal and vertical 
dimensions. With the 8-mc video band, 
selected for an example, we can deter- 
mine the resolution capabilities of such a 
system. 

It is to be noted that the present 
4.25-mc broadcast standard permits a 
balanced horizontal and vertical resolu- 
tion of approximately 400 television 
lines. An increase in the number of 
scanning lines, while retaining the 
bandwidth, reduces the horizontal de- 
tail as shown in Table I. The data indi- 
cate that an 8-mc system will give a 
balanced resolution when using 625 



Table I. Television Resolution.* 



Vertical 

Horizontal, Bandwidth 



4.25 me 
8 me 



525 625 735 819 

488 582 683 762 

340 283 240 216 

640 533 453 407 



*Scanning lines/60 fields, interlaced. 

318 March 1951 Journal of the SMPTE Vol. 56 



lines. The RCA PT-100 equipment 
has been designed to utilize the capa- 
bilities of a full 8-mc video channel. 

The signal-to-noise ratio is an ex- 
tremely important factor, probably the 
most important, if emphasis is to be 
placed on any one item. Motion picture 
film noise level, which until recently has 
not been quantitatively measured, 
should be the basis for an acceptable 
noise figure ; the value of 42 db for the 
electrical maximum signal-to-noise has 
been suggested. The type of noise is 
important, as impulse type can be ex- 
tremely troublesome because of its 
effect on the keyed d-c setting circuits. 
Also single-frequency noise will beat 
with the scanning frequency to form 
interference patterns which can be 
noticeable even though low in level 

The fourth factor, tonal rendition, is 
also important and is dependent on the 
operating conditions of the camera in 
particular, and on the operation of the 
projector as a secondary effect. 

When the camera characteristics can 
be well enough standardized, correction 
circuits can be introduced to enable the 
projector to produce pictures of photo- 
graphic reproduction qualities. The 
SMPTE Committees on theater tele- 
vision are studying these problems and 
will be able to make recommendations 
to the industry for the necessary stand- 
ards. 

Details of the PT-100 Equipments 

The design of the PT-100 projector is 
predicated on the choice of two elements 
of the projector: the kinescope and the 
optical system. The projection kine- 
scope chosen for the design was a 7-in. 
tube to be operated at 80,000 v; it was 
desirable to choose the smallest tube con- 
sistent with high performance in light 
output, in resolution, and in detail con- 
trast. 

With the kinescope size chosen, the 
RCA Tube Division undertook the task 
of developing the 7NP4 to fill the needs 
of this design, which is described in de- 



tail in an accompanying paper in this 
issue of the JOURNAL. 

Considerations of optical design re- 
quire a careful analysis because they are 
the most costly elements of the equip- 
ment. Their selection had to be pre- 
dicted on available manufacturing tech- 
niques for volume production of the 
glass blanks, the final grinding and the 
aluminizing. The cost of the optics and 
their mounting increases approximately 
as the square of the diameter. The 
production of the 42 in. mirror by 
laboratory methods was one thing on 
which cost was secondary, but the prac- 
tical considerations indicated that a 26- 
in. mirror would present a good com- 
promise. 

With the kinescope size chosen and 
the mirror size roughly determined, it 
was a problem of optics to arrive at the 
proper design center which includes the 
faceplate of the kinescope as an element 
of the optical system. 

The effective focal length finally 
chosen was 15.515 in., and Fig. 1 shows 
a chart of the operating conditions; the 
nominal projection throw is 62 ft from 
the face of the kinescope to the screen 
for a 20-ft wide picture. Since the 
"Schmidt" type system, as in the case 
of most other optics, is a fixed magnifi- 
cation device, the only variable is the 
picture size on the faceplate. Arbitrary 
limits are shown in the figure to in- 
dicate a degree of flexibility. The 
diagonals represent the diagonal ras- 
ter sizes on the kinescope, the nomi- 
nal size being 6}4 in. and shown as the 
center line. 

The percentage figures show the 
relative screen brightness for various 
conditions of operation with 100% for 
the nominal light output at the design 
center. 

With the limitation of the optics de- 
termined and the balcony location of the 
projector established, it was decided 
that a very minimum of equipment 
should be located in the theater audi- 
torium. 



Ralph V. Little, Jr. : Theater TV Equipment 



319 



26 



15.515 " E.F. SCHMIDT 



CORRECTOR PLATE FREE DIA.2l|' 
MIRROR FREE DIA.= 24-Z" 




60 70 

PRINCIPAL FOCUS TO SCREEN 



80 



Fig. 1. Chart of 15.515-in. effective focal length lens operation. 



Figure 2 gives a typical theater cross 
section showing how the projector would 
fit into a balcony-type house. The 
screen, due to the shallow depth of 
focus of the projector, is mounted nor- 
mal to the projection axis. The arrange- 
ment in a non-balcony or stadium house 
would require that the nominal throw be 
kept and that the projector be mounted 
on a retractable boom from the ceiling. 

The characteristics of the modula- 
tion, or video amplifier, made it neces- 
sary to place this element adjacent to the 
kinescope. In past designs it had also 
been necessary to include the horizontal 
scanning wave amplifier near the kine- 
scope deflection yoke, but this required 
more cabling to the projector and a 
great deal of physical space; therefore, 
a premise of the present design was to 
place all of the deflection equipment in 
the booth racks. The only electronic 
element of the equipment now remain- 
ing in the projector housing is the 
video power amplifier. The projector 



consists then of the projection kinescope 
(the 7NP4) the optical elements, a 26- 
in. mirror with a 22-in. correction lens 
together with the mounting or support 
of these elements. The electrical equip- 
ment is kept to a minimum with the re- 
quired video amplifier, a blower for 
cooling the kinescope faceplate, and the 
necessary terminal boards to facilitate 
interconnecting wiring. 

The equipment location can be seen in 
Fig. 3 which is a photograph of the pro- 
jector with one-half the outer housing 
removed. The wiring is accessible by 
lifting the top protective cover which re- 
veals the terminal board wiring side of 
the video amplifier. The amplifier is 
hinged to be tilted up and to the side, 
making the tube side of the chassis 
accessible and permitting adjustment or 
replacement of the kinescope. The pro- 
jector is interlocked through the control 
panel to remove the high voltage if the 
cover is raised ; the interlock also actu- 
ates a shorting arm which contacts the 



320 



March 1951 Journal of the SMPTE Vol. 56 




Ralph V. Little, Jr. ; Theater TV Equipment 



321 



high-voltage feed through bushing con- 
necting the circuit to ground. A lock on 
the cover gives added safety so that un- 
authorized persons cannot tamper with 
the equipment. 

High voltage is supplied from a unit 
designed to furnish the 80,000 v for the 
accelerating anode and also to furnish 
the focusing voltage of approximately 



18,000 v. The high-voltage supply is 
designed for remote operation and can 
be placed in a power or generator room 
near or adjacent to the projection booth. 
Two high-voltage cables lead from the 
supply directly to the projector and the 
control circuits are connected to the con- 
trol rack in the projection booth. 

The schematic diagram, Fig. 4, shows 




Fig. 3. PT-100 Projector, cover removed. 
March 1951 Journal of the SMPTE Vol. 56 



WL 2052 





WL 2052 



Fig. 4. High- voltage schematic, simplified. 



+ 80 K.V 



+ 40 K.V. 



FOCUSING 
-H8 K.V. VOLTAGE 



RCA-5890 



CONTROL 
VOLTAGE 



the elements of the supply consisting of a 
40,000-v transformer in a voltage- 
doubling rectifier circuit; the rectifier 
tubes are type WL2520 tubes. A special 
feature of this supply is the shunt regula- 
tor tube developed for remote adjust- 
ment of the focus voltage about its 
mean value of 18,000 v. 

The tube for focusing is the RCA 
5890; its use eliminates variable re- 
sistors with their attendant difficulty of 
insulation and stability at these high 
voltages. In addition to the basic ele- 
ments, the high-voltage supply contains 
protective circuits to short the output 
voltages when power is removed. 
Metering circuits are also provided with 
remote indication on the control panel to 
show the proper functioning of the 
equipment; the metering shows the 
voltage and current being developed by 
the supply. In operation the meter 
gives knowledge of the proper function- 
ing of the power supply; the current 
reading indicates that the kinescope is 
active and drawing power from the 
supply. The voltage indication shows 
the operation of the step-starting timer 
and shows when full voltage has been 
applied to the projector. 



Figure 5, a photograph of the high- 
voltage supply, shows the unit with the 
rectifier tubes exposed for servicing. By 
loosening four nuts this panel may be 
raised to the position shown. Through 
very conservative design it is expected 
that the rectifier tubes will last from 
three to five years; in fact, the entire 
unit will give years of uninterrupted 
service with a minimum of servicing. 
The only service function consists of 
rotating the rectifier tubes at stated in- 
tervals to keep the spare tube properly 
activated. 

Mechanical and Electrical 
Considerations 

The block diagram, Fig. 6, shows the 
location of the various parts of the sys- 
tem. There are the three logical di- 
visions of the equipment with their re- 
spective locations: the projector located 
in the theater, the projector control in 
the projection booth, and the high- 
voltage supply in the power or generator 
room. 

The location of the operating equip- 
ment in the projection booth gives pre- 
cedence for equipment design to con- 
form to time-tested procedure of front 



Ralph V. Little, Jr.: Theater TV Equipment 



323 




Fig. 5. High-voltage supply, tube-shelf open. 



PROJECTOR CONTROL 



PROJECTOR 



PROJECTION BOOT 
Off -AIR LINE 1 


H 






B 


M. 


CONY FACE 
.^^^^~ 

ff^ 


RECEIVER L 
MICROWAVE UNE g r 
COAXIAL CABLE 

r 

PRIMARY POWER 
1 1 7 V AC 2 K W ' 

L. 


SIGNAL 
SELECTOR 


CONTROL 
PANEL 8 

OSCILLOSCOPE 




VIDEO 
AMPLIFIER 


~ 






1 
L 




J 




L_ 

r- 

i 

L 










300V 
REGULATOR 


- VERTICAL 
-DEFLECTION 




J 


+ 18 K V 


+ 80 K V 

1 n 


1 
1 




400V. 8OOM A 
POWER 
SUPPLY 


- HORIZONTAL 
_ DEFLECTION 




| 
1 
1 








REGULATOR 
TRANSFORMER 


400V 40OMA 
- POWER 
SUPPLY 












HIGH VOLTAGE SUPPLY 
POWER ROOM 


'OFF THE AIR 
CHANNELS 2-13 

[-' 
|_ 


| PICTURE 
1 MONITOR 
L__,__J 

J 





324 



Fig. 6. PT-100; block diagram. 
March 1951 Journal of the SMPTE Vol. 56 



servicing of the equipment for theater 
work. The major electrical considera- 
tions, of course, are the Underwriter's 
requirements, and in addition, the ut- 
most in component reliability due to the 
economic necessity of installing single- 
channel equipment with no stand-by or 
emergency service available. 

The booth equipment consists of two 
short racks, one the projector control, 
the other the monitor rack. There are 
ten major units of equipment as shown 
by their respective blocks, and in ad- 
dition, the necessary terminal boards 
and a high-voltage control panel. 

The signal would enter the signal 
selector, which contains the video ampli- 
fier and synchronizing circuits, to be 
distributed to the control panel, the 
vertical deflection, and horizontal de- 
flection units. An 800-ma, 400-v 
power supply makes regulated 300 v 
available through the regulator unit. The 
400-ma, 400-v power supply furnishes 
power for the horizontal deflection and is 
fed from a 1-kw, regulating-type trans- 
former which also provides a standard of 
reference for the high voltage. An off- 
the-air receiver and picture monitor 
complete the complement of the racks. 



A typical chassis unit is shown in Fig. 
7, as it is mounted in the rack and 
ready for operation. A removable cover 
can be taken off to check the tubes or the 
fuses. All individual chassis have the 
primary power fused, as are the plate 
voltages which have neon indicators on 
their circuits. The unit shown is the 
vertical deflection amplifier; Fig. 8 
shows the interior exposed for servicing 
or adjustment of the infrequently used 
internal controls: the vertical hold, 
vertical size, vertical linearity and, on 
the rear panel, the vertical centering. 

Miniature tubes are used whenever 
possible and, in order to avoid mounting 
of parts on those small tube sockets and 
to make the unit more accessible for 
manufacture and servicing, resistor 
boards are used to produce this trim 
design. The mechanical design was so 
proportioned to permit standardization 
of the chassis blanks and the covers; 
in addition, it presents a uniform over- 
all appearance. 

Circuit Operation 

Electrically, the protection of the 
7NP4 kinescope required a major part 
of the design effort. The expense of the 




Fig. 7. Vertical deflection chassis, without cover. 
Ralph V. Little, Jr.: Theater TV Equipment 



325 



kinescope and the necessity of ensuring 
that the tube fulfill its life expectancy 
made electronic protection a must item. 
Scanning failure could cause the face- 
plate of the tube to be burned so as to 
make it unusable and in the extreme 
case, with high-voltage beam concen- 
trated on a single spot, that is without 
either vertical or horizontal scanning, 
it could melt a hole in the faceplate of 
the tube. 

Protection to the kinescope has been 
provided for the following contingen- 
cies: 

1. Open or shorted deflecting yoke, 

2. Lack of drive due to tubes, 

3. Loss of supply voltages, 

4. Overdrive of kinescope (positive 

grid). 

In operation, primary power is first 
applied for all electron-tube heaters and 
to the bias supply. A relay on the bias 
supply then closes, connecting primary 
power to the plate power supplies. 



Scanning will now be generated, but a 
series of interlocks must now be closed 
before the high voltage can be applied. 

The following protection circuits must 
be functioned properly: the horizontal 
scanning; the vertical scanning; the 
electromechanical interlock on the pro- 
jector cover must be closed; the vault 
high-voltage access door must be closed; 
then the high-voltage control circuit 
may be actuated. During operation ex- 
cess video drive could damage the kine- 
scope if it were not protected by an in- 
stantaneous bias control. 

The most important consideration in 
the design of the protection system is the 
speed with which failure can be de- 
tected and corrective measures taken 
and, in addition, the circuits must 'fail 
safely. The circuits operate in such a 
manner as to drive the kinescope to 
beam cut-off (and this must be ac- 
complished in a matter of less than 50 
n sec) and then the relays operate to re- 
move the high-voltage power. 




326 



Fig. 8. Vertical deflection chassis, inside. 
March 1951 Journal of the SMPTE Vol. 56 



MICROWAVE"! 



COAXIAL 
CABLE J 



TELEPHONE! , 

CONNECTION LIN 



VIDEO 

PROJECTOR MONITOR OSC LLOSCOPE 

2 2 PROJ 2 PROJ 



SYNC CONTROL 



AUDIO 



PROJECTOR MONITOR 

2 1 2 




AUDIO 



OPERATE CALIBRATE 

{K) c~-yi v 



3" OSCILLOSCOPE 




AUDIO MONITOR 
(HEADPHONES) 



THEATRE 
SOUND SYSTEM 



Fig. 9. Signal selector; simplified switching schematic. 




Fig. 10. Signal selector unit. 
Ralph V. Little, Jr.: Theater TV Equipment 



327 



Equipment Operation 

We must consider how the equipment 
will be operated in the projection booth, 
how the incoming signal will be con- 
trolled, and what points of operation 
should be checked. A signal selector 
panel was designed to facilitate the com- 
plete checking of the equipment prior to 
projection of the picture to the screen. 
Experience gained had shown the value 
of a system of checks to be made by the 
operator prior to show time. The 
switching will take care of two incoming 
lines each of video and audio signals, 
and also monitor the projector before 
the high-voltage power is applied. 

An off-the-air receiver is provided as a 
signal source during the initial period of 
use, or as a source of test signal if the 
normal signal is to come via microwaves 
or coaxial cable. The receiver is 
normally connected to Line 1 and an 
alternate signal is connected to Line 2. 
As auxiliaries to the signal selector a 7- 
in. picture monitor and a 3-in. oscillo- 
scope are used to check the projector 
functions without projecting a picture 



on the screen. The switching system 
of the signal is shown schematically in 
Fig. 9. It provides for the switching of 
the video and audio lines to the projec- 
tor and to the theater sound system, re- 
spectively, as well as to an oscilloscope 
and a monitor which are provided for 
level setting and quality control. 

The projector video amplifier has a 
cathode-follower video return which sup- 
plies a signal, attenuated by a ratio of 
100:1 from the kinescope drive. The 
return signal is marked Projector on the 
signal selector, and the incoming signals 
marked Line 1 and Line 2. 

The oscilloscope is a 3RP1 provided 
with a 60-cycle sine-wave sweep, d-c 
setting for the vertical deflection, and a 
calibration circuit set to provide 1 v, 
peak-to-peak, marker lines when its 
switch is set on calibrate position. 

In operation, Lines 1 and 2 are ad- 
justed for level using the oscilloscope, 
and then, when normal level is provided 
to the kinescope, by operation of the 
video attenuator the level for the projec- 
tor can be set by adjustment of the re- 
turn line from the projector. 




328 



Fig. 11. Control panel. 
March 1951 Journal of the SMPTE Vol. 56 



Fig. 12. PT-100 control and monitor racks. 
Ralph V. Little, Jr. : Theater TV Equipment 



329 



The 7-in. monitor can likewise be 
switched to view the pictures on the in- 
coming lines or from the kinescope in 
the projector; when on the projector, a 
complete system-operation check is ob- 
tained without requiring the picture on 
the theater screen. 

The monitor is provided with driving 
pulses from the projector scanning cir- 
cuits; it then shows the operation of the 
scanning lock-in as well as the picture 
quality; otherwise the monitor is syn- 
chronized from the incoming signal. 
The power supply for the monitor is 
self-contained making it independent of 
the operation of the projector. 

The audio signal can be obtained 
from either Line 1, which is the off-the- 
air receiver, or Line 2 which may be a 
telephone line connection. Projector 
audio would normally be connected to 
the theater motion picture sound system 
and is provided with an attenuator for 
level setting. In order to be able to 
check the presence of incoming signal 
before operations, a preamplifier and 
headphone circuit are provided with its 
own switching circuit, audio monitor 
Line 1 Line 2. When the audio has 



been switched to the theater sound cir- 
cuit, the normal monitor speaker will be 
in operation. Figure 10 shows the signal 
selector unit and the pushbutton con- 
trols are clearly identified with their re- 
spective functions. 

The control panel is the focal point of 
the operation and to fill the requirement 
of ease of operation, the controls were 
kept to the bare essentials. The panel 
contains: (1) the a-c control elements, 
the power and high-voltage on-off 
switch buttons; (2) a 3-in. oscilloscope 
for level setting; (3) a meter to indicate 
operation of the high-voltage supply and 
kinescope; (4) the operating controls 
for the video and audio; and (5) inter- 
lock indicator. 

For the operation of the equipment 
the functions of the control are shown in 
Fig. 11, with each control marked. To 
place the equipment in operation, it is 
only necessary to apply power by the 
power-on button; then to connect the 
equipment to the incoming signal and 
perform checks which are possible 
through the operation of the signal 
selector in conjunction with the monitor 
and oscilloscope. 




330 



Fig. 13. Projector. 
March 1951 Journal of the SMPTE Vol. 56 



Figure 12 shows the projection booth 
equipment consisting of the projector 
control rack on the right and the monitor 
rack on the left. The units of equip- 
ment from top to bottom are: the pro- 
jector control, the signal selector, the 
horizontal deflection amplifier, the 300- 
v regulator, and the high-voltage con- 
trol panel. The picture monitor is the 
top unit in its rack; below the receiver 
for off-the-air reception, the vertical de- 
flection amplifier, the 400-v, 400-ma 
power supply and the 400-v, 800-ma 
power supply. Several of the units of 
this equipment have not been described 
because of their conventional design. 

The projector is shown in Fig. 13, an 
external view of the completed unit 
ready for installation on the theater 
balcony face. 

Theater television as an entertain- 
ment medium has now reached the 
stage where a practical commercial 
equipment has been produced. To date, 
nine installations of the PT-100 equip- 
ment are in operation. The problems 
now he with the industry in the field 
application: the best way to transmit 
the program to the theater, the type of 



programming best suited for this new 
medium, and, most important in the 
long view, how will picture quality best 
be maintained from the program subject 
through the long chain of electronic 
events before the picture image is 
viewed on the theater screen? Every ef- 
fort, in the design of the PT-100 equip- 
ment, has been concentrated on building 
a quality product engineered for pres- 
ent-day use, and providing for standards 
which may be expected in the future as 
the needs of the industry are crystallized 
through the SMPTE. 

Bibliography 

I. G. Maloff and D. W. Epstein, "Reflec- 
tive optics in projection television," 
Electronics, vol. 17, pp. 98-105, Dec. 
1944. 

0. H. Schade, "Electrooptical characteris- 
tics of television systems," RCA Rev., 
vol. 9, nos. 1^, pp. 5-37, 245-286, 
490-530 and 653-686; Mar., June, 
Sept. and Dec. 1948. 

1. G. Maloff, "Optical problems in large- 
screen television," Jour. SMPE, vol. 
51, pp. 30-36, July 1948. 

R. V. Little, Jr., "Developments in large- 
screen television," Jour. SMPE, vol. 51, 
pp. 37-42, July 1948. 



Ralph V. Little, Jr.: Theater TV Equipment 



331 



Projection Kinescope 7NP4 
for Theater Television 

By L. E. Swedlund and C. W. Thierf elder 



The paper describes the design and development of the 7NP4, a 7-in ., 80-kv 
kinescope which is capable of providing clear, bright, theater-size (15 X 
20 ft) television pictures. The development involved solving design prob- 
lems including: (1) a high-efficiency, low-color-shift, white fluorescent 
screen; (2) adequate high- voltage insulation; and (3) a gun to provide 
electrostatic focus of a high-current beam into a small, sharp spot, and 
magnetic deflection through a relatively narrow angle to conserve deflec- 
tion power and provide essentially uniform focus over the entire picture 
area. 



THE 7NP4 is a high-voltage, projec- 
tion-type kinescope which provides 
a large, clear, theater-size picture when 
operated with a suitable reflective opti- 
cal system. The primary goal in the 
design of this kinescope was to provide 
a very bright fluorescent image of ade- 
quate resolution and contrast. In order 
to obtain an optimum design, virtually 
every element of the kinescope was in- 
vestigated and improved. 

At the beginning of the development 
of projection-type television in 1930, 
the light output from fluorescent screens 
was far from adequate for large pro- 
jected pictures, even with low resolution 
standards. Consequently, methods 



Presented on October 20, 1950, at the 
Society's Convention at Lake Placid, 
N.Y., by L. E. Swedlund and C. W. 
Thierf elder, Tube Dept., Radio Corpora- 
tion of America, Lancaster, Pa. 



other than direct projection from 
cathode-ray tubes having fluorescent 
screens were investigated. These alter- 
native methods fall into two classes: 
one using a cathode-ray tube having an 
incandescent or "cando-luminescent" 
screen, the other using a light-valve 
type of tube. l The former proved to be 
less efficient in producing light output 
than a fluorescent screen and also much 
more subject to overload damage. 
More success was attained with the 
light- valve method. This system de- 
pends on building up a temporary trans- 
parency that is intensely illuminated by 
a source of continuous light and is pro- 
jected on the viewing screen with a lens. 
In general, both designs suffer from low 
optical efficiency, low resolution and 
low contrast. The use of intermediate 
film, which also can be classed as a 
light-valve method, is perhaps the most 
successful of the group. Compared to 



332 



March 1951 Journal of the SMPTE Vol. 56 



direct projection from a kinescope hav- 
ing a fluorescent screen, the use of inter- 
mediate film has the disadvantages of 
time delay before projection, loss of 
resolution, the need for additional 
specialized operating personnel, and the 
high cost of film and processing. It has 
advantages, however, in that it can use 
standard movie projectors and provide 
a record for repeated showings. 

Optical projection of cathode-ray 
tube luminescent images received atten- 
tion from the advent of cathode-ray 
tube television, but initially the low 
tube-screen brightness, combined with 
low-efficiency optical systems, resulted 
in very dim images, even on small 
screens. The development of the 
Schmidt camera lens for cathode-ray 
tube projection, 2 giving an approxi- 
mately fivefold gain in optical efficiency 
over the best refractive optical system, 
marked a considerable advance toward 
the realization of theater-size television 
pictures. In 1941, a system of this 
type, utilizing a 73^-in. kinescope 
operating at 65 kv, was demonstrated 
in New York City. 3 This development 
was interrupted by the war, but since 
its resumption, continuous progress has 
been made, not only in increasing light 
output, but also in improving quality, 
reliability and tube life. 



Light-Output Factors 

The light output of a luminescent 
screen in a kinescope depends primarily 
on the energy of the electron beam and 
the light-conversion efficiency of the 
screen. The electron energy per unit 
time is the product of beam voltage and 
current. The maximum value of beam 
current at a given beam voltage is de- 
termined by the rate at which the energy 
of the electron beam can be absorbed 
by the phosphor, and by the maximum 
current which can be sharply focused 
by the electron gun. As a result, the 
beam current is limited to a few milli- 
amperes. The principal means of in- 



creasing the energy, therefore, is to 
raise the beam voltage. In addition 
to increasing the energy of the electron 
beam, a higher beam voltage improves 
the light-conversion efficiency of the 
luminescent screen and permits a higher- 
current electron beam to be sharply 
focused by the electron gun. Maximum 
light output, therefore, increases with 
voltage at a greater rate than indicated 
by a linear relationship bet ween the two. 
Raising the voltage, however, makes 
necessary the provision of better insu- 
lation and higher deflection energy; 
hence a voltage limit is introduced 
which depends on how well these re- 
quirements can be met. 

The efficiency of the conversion of 
electron beam energy into visible light 
is determined chiefly by the choice of 
materials used in the fluorescent screen. 
The density of the electron beam strik- 
ing these materials is also a factor in- 
fluencing the efficiency. 

High light output from the 7NP4 is 
obtained chiefly by using a high-effi- 
ciency screen and an 80-kv anode poten- 
tial. 

Size of 7NP4 Screen 

The light output of a projection 
kinescope can be increased somewhat if 
the size of the luminescent screen is in- 
creased. This method of raising the 
light output is of secondary importance, 
however, and is limited by the desirable 
and practical bulk and cost of the optical 
system. 

The increase in light output with in- 
creasing screen size is made possible 
by the attainment of higher conversion 
efficiency. The factors responsible for 
the variation of efficiency with varying 
screen size are: (1) the current density 
saturation of the luminescent screen ma- 
terial and (2) the temperature of the 
screen. The larger focused spot on a 
larger screen results in lower current 
density and hence in less saturation. 
The greater cooling surface and lower 



Swedlund and Thierfelder: Projection Kinescope 



333 



to 


ANOO 
ANOD 


VOLTS 

: CURR 


= 80000 
:NT=6 


MA 








u 

-1 
















r 










^ 


_ 




i 






/ 


^ 








X 

< 3000 
















5 300C 
2000 
















S 


i 7 9 II 13 1 





TUBE FACE DIAMETER-INCHES 



Fig. 1. Variation of maximum light 
output with changing screen size. 



power density permit operation at a 
lower screen temperature. Figure 1 
shows the variation in peak light output 
with screen size. 

Although an increase in screen dimen- 
sions does permit some increase in 
light-conversion efficiency, it does not 
permit appreciable increase in beam 
energy. It might seem that the use of 
a larger screen, by permitting the em- 
ployment of a larger spot for the same 
resolution, would permit the use of a 
higher beam current. But this is not 
so. Even when beam current is held 
constant, the spot size increases practi- 
cally in proportion with screen size, for 
a given deflecting angle and beam volt- 
age. Therefore, an increase in current 
would enlarge the spot size beyond the 
limit imposed by the requirement for 
good resolution. 

In spite of the fact that a larger 
screen results in somewhat greater light 
output, the size, weight and cost of the 
optical system increase so rapidly with 
increasing screen diameter that practi- 
cal considerations limit the size of the 
screen. The volume and weight of the 
optical system are approximately pro- 
portional to the cube of the kinescope 
diameter, and the cost of the optical 
system increases approximately as the 
square of the kinescope diameter. Pro- 
jection kinescopes with diameters of up 



to 15 in. have been experimentally 
made and tested. It is found that a 
kinescope with a faceplate diameter of 
7 in. gives a good balance between the 
light output and the bulk and cost of the 
reflective projector. 

Screen Materials 

A luminescent screen with a high 
efficiency in converting the electron 
beam energy into light was one of the 
goals in the development of the 7NP4. 
Sulfide-type phosphors are normally the 
most efficient, but, because of current- 
saturation effects, their efficiency is 
inferior at the high current density used 
in a projection tube. Silicate-type 
phosphors are less efficient at low cur- 
rent densities but, because they undergo 
less saturation, their over-all efficiency 
hi a projection tube is greater. They 
are, therefore, preferred for this applica- 
tion. There are relatively efficient 
silicates which emit in the yellow- 
orange spectrum, but unfortunately the 
blue-emitting silicate has rather low 
efficiency, although it has even less cur- 
rent saturation than the yellow-emitting 
material. If the screen is composed of 
a mixture of yellow-emitting silicate 
and blue-emitting sulfide, the efficiency 
will be relatively high but the color will 
shift toward the yellow as the current 
density is increased. This color shift 
is objectionable but can be reduced by 
settling the sulfide against the glass face- 
plate and the silicate in a layer next to 
it, exposed to the incident electron 
beam. Then, since the beam is partly 
absorbed and diffused while it is pene- 
trating the silicate layer, the current 
density is lower when the beam reaches 
the sulfide. But even with the reduc- 
tion in current density due to the rela- 
tively thick layer of screen material re- 
quired for efficient high-voltage opera- 
tion, the color shift is still objection- 
able. The color shift was finally re- 
duced to a negligible value by the sub- 
stitution of blue-emitting silicate-type 



334 



March 1951 Journal of the SMPTE Vol. 56 



phosphor for part of the blue-emitting 
sulfide phosphor. 

The optimum screen weight for the 
7NP4 is approximately 8 mg of phos- 
phor material per square centimeter. 
This value is about three times as great 
as that used in a low-voltage kinescope 
of the directly-viewed type. An im- 
portant feature is the aluminum surface 
applied to the back of the screen to 
maintain it at full operating potential 
and to reflect light from the back to the 
front. 4 

Insulation Considerations 

In the projection tube the external 
high-potential and low-potential ter- 
minals cannot be separated by the full 
length of the tube, as is done in X-ray 
tubes. A deflecting yoke which is 
approximately at cathode potential is 
fitted over the projection tube neck, 
near the midsection of the tube. There- 
fore, the full anode potential is applied 
over half the tube length. The yoke is 
grounded and the cathode should be 
operated at or near ground, rather than 
negative to ground as in X-ray tubes, 
in order to facilitate the design and pro- 
duction of equipment using the 7NP4. 

Adequate air insulation is obtained 
for the projection tube by placing the 
anode terminal as near the face of the 
tube as possible and by selecting a nar- 
row deflection angle to make the cone 
long. The surface leakage path is in- 
creased by molding circumferential cor- 
rugations in the cone section of the bulb. 
In order to avoid the formation of a 
film of moisture on the glass, which 
tends to promote corona and arcing, 
the cone is coated with a moisture- 
repellent, insulating lacquer coating. 
This coating is colored black to help 
reduce stray light reflections in the 
optical barrel (Fig. 2). 

The air space between the yoke and 
the neck may be another source of 
corona and breakdown because it is a 
dielectric in series with the glass neck, 
and the full anode voltage is applied 





Fig. 2. Photograph of 7NP4 kinescope. 



across the two. Due to the fact that 
the air has a much lower dielectric con- 
stant than the glass, it is subjected to a 
higher voltage stress and may break 
down. Such breakdown would result 
in corona, heating and possible failure 
of the glass insulation. This voltage 
stress is avoided by applying a conduct- 
ing lacquer to this part of the neck and 
connecting it to ground. 

Internally the cone is at anode poten- 
tial from the luminescent screen to the 
anode end of the electron gun (Fig. 3). 



Swedlund and Thierf elder: Projection Kinescope 



335 




\ \\_GRID 
PLASTIC\ N2 2 
FILLED ^CATHODE 



DEFLECTING 
YOKE 



Fig. 3. Cross section of 7NP4 kinescope. 



SPHERICAL 
MIRROR 



CORRECTING LENS 
& X-RAY SHIELD 

Wl 



. n7 DIA FACE PLATE 

/ T 7 DIA - OF 7NP4 

BLOWER |T"t 
7"DIA. 




2l'/ 2 

WORKING 
DIA. 




Fig. 4. Typical optical and 
cooling system for thea- 
ter - television projec- 
tor using 7NP4 kinescope. 



Thus, the full anode voltage is applied 
between the outer conducting lacquer 
and the inside of the neck under the 
yoke. In a conventional design, there- 
fore, the glass wall of the neck of the 
tube would be stressed by the full 80- 
kv potential difference. But it is 
difficult to make a glass wall uniform 
enough to withstand more than 40 kv. 
The insulation problem posed by this 
property of glass was solved by D. W. 
Epstein who placed a second, larger 
neck around the portion of the neck 
carrying the high anode voltage and 
provided vacuum insulation between 
the two necks. 
Another insulation problem was to 



provide an external terminal for the 
focusing electrode, which operates at 
17 kv. A first attempt at solving this 
problem was made by sealing a terminal 
in the glass neck. This terminal, how- 
ever, had to be flush with the surface in 
order to provide clearance for the yoke. 
A better method of making contact was 
sought in order to obtain a trouble-free 
connection and to facilitate insertion 
and removal of the tube from the pro- 
jector. After various materials and 
structures for the stem leads and base 
insulation were investigated, it was 
found that if a plastic material having 
high dielectric strength replaced the 
air in the space inside the base, the 



336 



March 1951 Journal of the SMPTE Vol. 56 



focusing voltage connection could be 
made through the base. A novel pro- 
cedure developed for producing this 
structure consists of filling the base 
with an exothermic plastic mixture and 
polymerizing it in place. The plastic 
also serves to cement the base to the 
neck. 

Faceplate Considerations 

The faceplate of the 7NP4 projection 
kinescope is designed to be used in a 
Schmidt-type reflective optical system 
(Fig. 4) and as such has to be a precision 
optical element accurately aligned with 
the rest of the system. The faceplate 
is a spherical section with a radius of 
curvature of 15.315 in. as required by 
the design of the optical system. It is 
not difficult to grind and polish the face- 
plate glass to the desired optical toler- 
ance, but unless special techniques are 
used when it is being sealed to the cone 
of the bulb, it is distorted much more 
than can be tolerated. It was found 
desirable to grind the glass to a radius 
of curvature slightly greater than that 
required by the optical system and to 
allow the radius to decrease during the 
controlled sealing and glass-annealing 
operations. The glass face is accurately 
aligned with the bulb neck in order to 
minimize the amount of adjustment 
needed in the projector. 

It has been known for many years 
that glass darkens when subjected to 
high-voltage electron and X-ray bom- 
bardment. This phenomenon is readily 
observed in X-ray tubes, but it is not 
objectionable there because it does not 
affect X-ray transmission. The color 
of the darkening is usually a yellowish 
brown and its intensity varies con- 
siderably with the glass composition. 
The soft glasses, such as ordinary 
window glass, darken much more 
quickly than the hard or borosilicate 
glasses and cannot be cleared or bleached 
appreciably by heating as can many of 
the latter types. The darkening ap- 
pears to be due partly to electron bom- 



bardment of the glass, which affects 
the surface, and partly to X-ray bom- 
bardment, which produces a darkening 
that extends into the body of the glass. 
The darkening caused by X-rays can 
be bleached by heating the glass to 
about 200 C with infrared radiation. 

Corning Type 774 Pyrex glass, a 
hard glass commonly used for laboratory 
and high-voltage glassware, darkens 
objectionably in about 50 hr of operation 
when used in an 80-kv theater-projec- 
tion kinescope. Extensive tests over a 
long period of time were run in co- 
operation with the glass companies to 
find or develop a glass which would 
evidence little or no darkening. Al- 
though a few were found, they had 
other characteristics which made them 
unsuitable. Finally, methods were de- 
vised for overcoming the optical de- 
ficiencies of one of these types having 
little darkening, and this glass, Corning 
Type 707, was chosen for the 7NP4. 
This glass is a low thermal-expansion, 
hard glass, originally developed to have 
good high-voltage and electrical loss 
characteristics. Until recently, how- 
ever, this glass had not been produced 
with satisfactory optical quality. But 
when special care is exercised and 
larger melts are made, 707 glass of 
relatively good optical quality is ob- 
tainable. A few seeds or small bubbles 
have to be tolerated but the faceplates 
are selected so that these blemishes are 
not near the inside surface. Because 
the optical system has a very shallow 
depth of focus, these imperfections are 
not noticeable in the image. This glass 
does darken gradually, but it can be 
bleached by heating with infrared radia- 
tion lamps for about 20 min as shown in 
Fig. 5. This treatment should not be 
needed before about 150 hr of operation. 

Electron Gun Considerations 

The electron gun must project a very 
intense electron image on the lumines- 
cent screen. The spot diameter at 
maximum current should not be much 



Swedlund and Thierf elder: Projection Kinescope 



337 



TOP VIEW 




LATERAL VIEW 

AT AA' OMITTING 

LAMPS I&3 



Fig. 5. Arrangement of infrared 
lamps for bleaching 7NP4 faceplate. 



larger than the line width of a 525-line 
image on this tube (7.5 mils). The 
spot size will decrease with current, a 
tendency which helps to improve 
shadow detail. Because, on the other 
hand, deflection acts to increase spot 
size, some allowance has to be made for 
this effect. The electron beam can be 
focused either with an external magnetic 
field or an internal electrostatic field. 
With good design, both systems can be 
made to function with negligible aberra- 
tion; the choice, therefore, can be 
made on the basis of ease of manufacture 
and ease of mounting and operating 
the tube. 

In a reflective optical system, the 
kinescope and its accessories block out 



reflected light. For this type of appli- 
cation, therefore, electrostatic focusing 
is preferable because it results in more 
compact kinescope accessories and re- 
moves the problem of providing sup- 
port and adjustments for a relatively 
heavy magnetic focusing coil. Elec- 
trostatic focusing voltage may be ob- 
tained from a bleeder resistance across 
part of the anode-voltage supply. This 
method is advantageous because it pro- 
vides automatic correction of focus. 
As the brightness of the image varies, 
the load on the voltage-anode supply 
changes, causing the voltage applied 
to the focusing electrode to vary. A 
change in focusing-electrode voltage 
occasioned, for instance, by a switch 
from a dark to a bright picture, results 
in a change of focus such that a sharp 
picture is maintained. In contrast, 
with magnetic focus the focusing field 
is manually adjusted and is not readily 
controlled automatically because of the 
inductance of the focusing coil. 

A further reason for using electro- 
static focusing in the 7NP4 is that this 
type of focusing reduces an insulation 
problem. The voltage across the gap 
between the anode and the adjacent 
focusing electrode is about 63 kv. 
With magnetic focusing, the voltage 
across the gap between anode and the 
adjacent point of the electron gun would 
be 80 kv. An additional advantage is 
that, when the RCA 5890, a new high- 
voltage regulator tube, is used, one may 
adjust the focusing voltage with a low- 
voltage, low-current, remote-control 
potentiometer. A disadvantage of elec- 
trostatic focus compared with magnetic 
focus is that it increases the difficulty 
of tube manufacture, because the gun is 
a little more complex. Nearly per- 
fectly circular electrodes must be pro- 
vided to produce the symmetrical focus- 
ing field required for a round spot. 

The electron gun in this tube has an 
oxide-coated cathode. The voltage rat- 
ing of the gun is probably higher than 
that of any other vacuum-tube unit 



338 



March 1951 Journal of the SMPTE Vol. 56 



140 



150^ 



100 90 80 70 



TYPE 7NP4 

ANODE VOLTS = 80000 



MEASURED AT DISTANCE 
OF 15 INCHES FROM 
CENTER OF FACE 



10 20 30 4O 5O 6O 7O 

X-RAY INTENSITY 
ROENTGENS/HOUR/ MA 




Fig. 6. Polar distribution of X-Ray radiation from 7NP4. 



with an oxide-coated cathode. An 
oxide-coated cathode is desirable be- 
cause its high emission permits forma- 
tion of a high-current-density beam 
and thus helps provide sharp focus. 
Furthermore its low work function 
greatly reduces the control voltage re- 
quired. Reliable operation of an oxide- 
coated cathode at very high voltage is 
made possible by means of excellent 
exhaust and getter systems which pro- 
duce and maintain a very low gas pres- 
sure and limit ion bombardment of the 
cathode surface. The getter is flashed 
over the entire inside surface of the cone, 
providing a relatively large active area 
close to the main sources of ionized gas 
molecules. A getter flash over the 
inside of the bulb is possible because the 
aluminum screen backing protects the 
luminescent screen and because a de- 
posit of getter on the screen does not 
appreciably reduce the energy of the 
high-voltage beam. 

Deflection Considerations 

Magnetic, rather than electrostatic, 



deflection is always used for high-voltage 
cathode-ray tubes to avoid insulation 
and distortion problems inherent with 
the latter. 

A smaller deflection angle than that 
used in kinescopes for home-television 
receivers was chosen for the 7NP4 in 
order to reduce deflection-power re- 
quirements and to minimize loss of edge 
resolution. The larger-diameter neck 
required for the double-neck insulation 
makes deflection more difficult because 
the length of the deflecting yoke field is 
determined by the clearance of the inside 
neck. The deflecting power needed is 
of reasonable proportion, however, being 
not much greater than that required for 
wide deflection-angle, home-television 
receivers. A good margin of deflection 
power for reliability and good linearity 
is, therefore, readily provided. A nar- 
row deflection angle also requires an in- 
crease in the distance between the 
electron gun and the screen. The in- 
crease in distance increases spot mag- 
nification, but this effect is compensated 
for by making the electron gun longer. 



Swedlund and Thierf elder: Projection Kinescope 



339 



100 



80 



20 




3000 5000 7000 

WAVELENGTH -ANGSTROMS 

Fig. 7. Spectral-energy emission char- 
acteristic of phosphor No. 4, silica te- 
sulfide type; color temp., 6300 K. 



Application and Operation 

The 7NP4 operates at a voltage 
which produces penetrating X-rays. 
The radiation is emitted in a pattern 
that is symmetrical about the tube axis, 
as shown in Fig. 6. The X-ray in- 
tensity indicated is representative of 
normal operation. Although this level 
is low compared to that of a commercial 
X-ray tube, it is several thousand times 
the safe level for continuous exposure 
at close range. Sufficient shielding can 
readily be provided by enclosing the 
optical system in a metal barrel. Be- 
cause radiation may extend for several 
feet in front of the projector, an X-ray- 
absorbing window may be needed. 

The screen emits white light which has 
an equivalent black-body color tempera- 
ture of approximately 6300 K. The 
spectral distribution of the energy 
emitted by the screen is shown in Fig. 7. 
At peak light output the maximum 
power input to the screen is approxi- 
mately 480 w. For this condition the 
peak power in the beam is approximately 
1000 kw/sq cm. This power density is 
extremely high and it is only because 
the spot moves across the screen at high 
velocity that the screen is not destroyed. 
Hence it is extremely important to pro- 



TYPE 7NP4 
_E.p = 6.6 VOLTS RASTER SIZE 5" 3 V 

CATHODE-DRIVE SERVICE: P-G| VOLTS 
= 750OO; G3~G| VOLTS ADJUSTED 
TO FOCUS; G2~G| VOLTS ADJUSTED 
TO PATTERN CUTOFF; K-G| VOLTS 
= I25(K+) 

GRID-DRIVE SERVICE: Eb= 75000 V ; 
G 3 VOLTS ADJUSTED TO FOCUS; G 2 
VOLTS ADJUSTED TO PATTERN 
CUTOFF; EC, =- 155V. 




VIDEO SIGNAL VOLTS FROM CUTOFF 



Fig. 8. Average drive characteristics 
of 7NP4. 

vide full scanning at all times when the 
beam is on. 

The faceplate of the 7NP4 is an optical 
component and has to be accurately 
positioned in the optical system. The 
high-voltage anode terminal is near the 
faceplate, so a support here to position 
the faceplate would require full anode- 
voltage insulation. The deflecting yoke 
is at ground potential and is thus a 
more convenient place to support this 
tube. Due to the method of assembly 
of the glass bulb parts, the neck is well 
aligned with the faceplate, but lateral as 
well as longitudinal adjustments are 
needed to position and focus the lumines- 
cent screen properly in the optical 
system. 

In normal operation 80 to 160 w may 
be dissipated at the screen. With only 
radiation and convection cooling, this 
dissipated power would produce an ex- 
cessive temperature and thus reduce the 
efficiency of the luminescent screen and 
possibly fracture the glass. A stream 
of air from a small blower, directed 
perpendicularly at the faceplate through 
a hole in the center of the mirror, is 
sufficient to keep the glass temperature 



340 



March 1951 Journal of the SMPTE Vol. 56 



30000 

to 

H 20000 



2 

< 10000 
*_ 8000 

8 600 

1 ^OOO 

2 3000 

z 

t 2000 



1000 
800 

600 



400 
300 



34 6 8 10 



TYPE 7NP4 
E.f= 6.6 VOLTS RASTER SIZE 5 X 3 3 4 
CATHODE-DRIVE SERVICE: P~G| VOLTS = 75000; 
G3~G| VOLTS ADJUSTED TO FOCUS; G2~G| 
VOLTS ADJUSTED TO PATTERN CUTOFF; 
K-G| VOLTS = 125 (K+) 
-GRID-DRIVE SERVICE: Eb= 75000 V.; 03 VOLTS 
ADJUSTED TO FOCUS; G? VOLTS ADJUSTED 
TO PATTERN CUTOFF; EC| =- 155 V. 








II 
TRANSFI 
TRANSF 
ON BA 


:R c 

ER C 
SIS 
OF 


HA 

:HA 

OF 
IOC 


RA< 
RA 
CO 

>:| 


i 

:TERU 

ITERli 

NTRA: 


I 

5TICS 

















5TICS 
5T 












^ RATIO 





























i 


~I /$ 
1 /<$ 

IQ 

/ft- 




















- 


i<3 












































b 
























/ 






















/ 


/ 






















/ 


n 






















V 






















/ 


/ 




















t 


/; 




















/ 


I f 


f 




















/ 


* 






















/ I 


























1 




















/ 


J 


1 




















s 


SI 


/ 


















. 


''**' 


1 






1 


1 







20 



40 60 80 100 



VIDEO SIGNAL VOLTS FROM CUTOFF 
Fig. 9. Average drive characteristics of 7NP4 in log-log coordinates. 



below 100 C, the recommended maxi- 
mum temperature for the outside of the 
face glass. The cooling air should be 
filtered to eliminate dust and minimize 
moisture formation inside the optical 
barrel. 

Because of the relatively high beam 
current desired, an average grid-drive 
voltage of 155 v is needed. Inasmuch 
as a generous video bandwidth is needed 
to produce good detail, it is desirable 
to reduce this voltage requirement if 
possible. A worth-while gain is ob- 
tained by applying the video signal to 
the cathode instead of to grid No. 1. 
For example, with grid No. 1 grounded 
and the signal applied to drive the 
cathode negative, the grid No. 1-to- 
cathode voltage decreases and in addi- 
tion, the grid No. 2-to-cathode voltage 



increases (Fig. 8). Both actions in- 
crease the beam current. The signal 
required is thereby reduced to 125 v by 
the compound action of the drive 
voltage. 

Figure 9 shows the transfer character- 
istics on a log-log graph in the form used 
to indicate photographic film character- 
istics. This form of presentation has 
been described by 0. H. Schade. 5 The 
transfer characteristic is obtained by 
measuring the brightness of an un- 
modulated raster. In an actual picture 
some light is scattered into the black 
areas. This effect will vary with the 
type of picture, but 1% of scattering is 
a representative value. The result is 
a contrast ratio of 100:1 in the lumines- 
cent screen. A second transfer char- 
acteristic, taking this reduction of con- 



Swedlund and Thierfelder: Projection Kinescope 



341 



trast into account, is shown as a dotted 
line. The slope or gamma of the 
straight part of these characteristics is 
2.4 and 2.6 for cathode drive and grid 
drive, respectively. It is not possible to 
vary this slope appreciably by gun de- 
sign changes. The gamma of the signal 
can be modified in the video amplifier 
and, of course, should be adjusted to 
provide the desired over-all gamma. 

The expected life of a theater projec- 
tion kinescope is an important character- 
istic. Although there is relatively little 
actual operating experience, simulated 
life tests indicate that an average life 
expectancy of 500 hr is possible. 



References 

1. See "Theater Television Bibliography," 
Jour. SMPE, vol. 52, pp. 268-272, 
Mar. 1949. 

2. U.S. Pat. 2,273,801, Feb. 17, 1942. 

3. RCA Rev., vol. 6, no. 1, pp. 5-11, 
July 1941. 

4. W. Epstein and L. Pensak, "Improved 
cathode-ray tubes with metal backed 
luminescent screens," RCA Rev., vol. 
7, pp. 5-11, Mar. 1946. 

5. O. H. Schade, "Electrooptical char- 
acteristics of television systems," RCA 
Rev., vol. 9, nos. 1-4, pp. 5-37, 245- 
286, 490-530 and 653-686; Mar., 
June, Sept. and Dec. 1948. 



342 



March 1951 Journal of the SMPTE Vol. 56 



Installation of Theater 
Television Equipment 

By E. Stanko and C. Y. Keen 



Recent initial installations of the RCA PT-100 Theater Television System 
are described. Problems encountered in these installations and their 
practical solution are discussed. Now that the first commercial Theater 
Television Systems have been installed in a number of representative 
theaters, the experience acquired at these installations will be valuable in 
reducing the installation of theater television equipment to a routine pro- 
cedure, comparable to the installation of sound motion picture equipment. 
Procedures followed in making these initial installations are described, 
beginning with the preliminary theater survey to the final installation 
check and adjustment before projecting a picture on the screen, including 
service and maintenance problems. 



Preliminary Theater Survey 

Before an installation of theater tele- 
vision equipment is made in any thea- 
ter, certain facts about the theater must 
be known. One of the most important 
items to be determined by the survey is 
the location of the equipment, especially 
the projector unit. The PT-100 Sys- 
tem Projector employs a Schmidt-Type 
Optical System with a correction lens 
which is limited in projection "throw" 
to a definite operating range. 

Figure 1 shows the operating range of 
the optical system. There are three 
variables : 

1. Picture size, 

2. Kinescope raster* size, and 

3. Projection throw from projector to 
screen. 



Presented on October 20, 1950, at the So- 
ciety's Convention at Lake Placid, N. Y., 
by E. Stanko and C. Y. Keen, RCA Serv- 
ice Co., Inc., Camden, N. J. 
*Raster is defined as the illumination 
caused by the scanning lines on the 
cathode-ray screen when no television 
picture signal is being received. Picture 



So far all of the initial installations 
have been made in balcony-type thea- 
ters, where the projector is mounted at 
the front face of the balcony or in the 
front part of the balcony structure. 

The preliminary survey includes an 
elevation outline plan of the theater 
with dimensions so that the three vari- 
ables shown on the chart can be resolved 
and the position of the projector and 
screen and screen size, determined. 

Theater plans are obtained where 
possible and are carefully examined so 
that the theater exhibitor can be ad- 
vised how to proceed with the installa- 
tion. 

The survey also determines the loca- 
tion of the control racks in the projec- 
tion booth and the location of the 80-kv 
high-voltage power supply unit, and in- 
cludes information that will be required 
to solve any unusual problems in con- 
nection with the installation of the 
equipment. 



size is slightly smaller due to blanking 
interval. 



March 1951 Journal of the SMPTE Vol. 56 



343 




60 70 

Projection Throw From Projector To Screen in Feet 
Figure 1. 



80 



Antenna. Surveys 

There are several means for bringing 
a television program to the theater. 
Video line facilities, consisting of coaxial 
cable or an equalized pair, is one method. 
Point-to-point transmission by micro- 
wave relay equipment is another 
method. "Off-the-air" pickup on the 
present television channels is still an- 
other method. 

The PT-100 equipment is designed to 
receive programs by any one or all of 
these methods. 

If "off-the-air" pickup is a require- 
ment, it may necessitate special an- 
tenna facilities. In this case, at the re- 
quest of the theater exhibitor, an an- 
tenna survey is made with the use of 
field-measuring equipment to deter- 
mine the possibilities of obtaining an 
acceptable "off-the-air" signal and pre- 
pare installation instructions and spe- 
cifications for an antenna installation 
suitable for an electrical contractor to 
use in order to insure getting the best 



possible signal obtainable at the theater 
location. "Off-the-air" program pickup 
is not generally considered to be an 
entirely satisfactory means of providing 
theater television programs, as the pic- 
ture quality obtainable is, in most cases, 
below the performance possible with the 
PT-100 equipment. Figure 2 shows an 
antenna survey crew ready to make a 
field strength survey. 

Installation of the PT-100 Equipment 

Installation of this equipment can be 
divided into five principal operations. 
(Figure 3 shows the PT-100 Exhibitors 
Installation Instructions.) 

1. Installation of the projector, 

2. Installation of projection-booth 

rack equipment, 

3. Installation of the high-voltage 

supply unit, 

4. Electrical connection of equipment 

units, 

5. Hanging the screen. 
Installation of the Projector (Figs. 



344 



March 1951 Journal of the SMPTE Vol. 56 



The PT-100 System Projector is sup- 
ported on legs at its center of gravity 
and can be pivoted in a vertical direc- 
tion 10. Four bolts are used in each 
supporting leg to secure the projector 
to a supporting plate or steel structure 
which must be provided by the theater. 
The projector weighs 400 Ib. The loca- 
tion of the projector, as determined by 
the preliminary theater survey and the 
construction of the theater, will deter- 
mine how it is to be mounted. 

In several of the initial installations 
of this equipment the projector was 
mounted with the legs horizontal. 
Several other installations were made 
with the legs mounted vertically. The 
method of mounting will vary in dif- 
ferent theaters although some degree of 
standardized practice may be evolved 
from the experience gained with the first 
installations. 

If the projector is mounted on the 



front face of the balcony it will require 
a supporting steel framework secured to 
the balcony steel structure with a flat 
surface to which the legs are bolted. 

When the legs are mounted in the 
horizontal position, care must be taken 
to be sure that the supporting structure 
allows for some adjustment in the hori- 
zontal plane to insure that the optical 
axis of the projector is centered on the 
screen which is centered with the theater 
center line. 

If the projector is mounted with the 
legs vertical it is usually easier to ad- 
just it so that the optical axis is cen- 
tered. 

One method of mounting the projec- 
tor that has some practical advantages 
utilizes two I beams projecting from the 
front of the balcony, which are secured 
to the balcony main I-beam truss and, 
further back, to the balcony steel struc- 
ture in a cantilever arrangement. 




Fig. 2. An antenna survey crew ready to make a field strength survey. 

Stanko and Keen: Installing Theater TV 345 



MRC-59/U f IF REQUIRED-* 

^(rrTrr, 1 I LV.'.fH 






9N0.14BRC 
4 RG-59/U 
I PR.19DLD 
I No. WOI7) 

SPECIAL CABLE 

MAKE NO 

SUBSTITUTION 



NOTE' INSULATION OF WIRES Mo .117 t 118 MUST 
BE NOT LE53 THAN I ; OOO-VOLT RATING 
X= SPABE CIRCUITS 



MONITOR RACK Mi-isooe 




rteJ-i rttnttgfntmtth pr: 

IN I I 2 3 4 S 6 7 6 9 Oil 12 13(415 16 17 I OND 



CONTROL PANEL 6T83 



SYNC AMPLIFIER 7TB2 



T ' ON HOR SWEEP 

OUT -^-I0< 

CHASSIS (OND 

CONTROL RACK MI-ISOO? 

PROJECTION ROOM 



Fig. 3. Exhibitor's installation 

The I beams are separated sufficiently ing structure does not interfere with the 



to allow the projector legs to be mounted 
vertically on a plate which is bolted 
across the bottom of the I beams at the 
required distance to provide about one 
foot of space between the rear of the 
projector and the balcony face. 

Provision must be made in any 
method of mounting so that the mount- 



high-voltage cable connection to the 
barrel or the wiring entering the top rear 
conduit plate. 

A servicing platform must be pro- 
vided on the right side of the projector, 
the floor of which is on a level with the 
bottom of the legs when they are in the 
vertical position. 



346 



March 1951 Journal of the SMPTE Vol. 56 



NOTE: TERMINATC 

CONDUIT WITH , 

LENGTH OF l-l/ 

TO ALLOW 3/4" GREENFIELD 

TO SLIDE INSIDE TO 



ENABLE MAKING COUPLING 

TO POWE 



THIS HIGH-VOLTAGE CABLE 
IS APPROXIMATELY l-l/" 
OUTSIDE DIAMETER OVER 
ITS COPPER BRAID SHIELD 
AND PROTECTIVE LOOM 
JACKET USE 2"CONDU(T 
OR OTHER PROTECTION- 
DEPENDING ON LOCAL 
REGULATIONS. 

ABLE IS/ 1 * 



THE RG-34/U CABLE IS 
5/8" OUTSIDE DIAMETER 
AND HAS A VINYL PRO- 
TECTIVE COVERING OVER 
ITS BRAIDED COPPER 
SHIELD USE 3/4"CONDUIT 
OR OTHER PROTECTION- 



POWER 
SUPPLY 

UNIT MI-15115 




instructions for the projector. 



The projection angle will vary in dif- 
ferent theaters but should be kept as 
small as possible. The screen must be 
angled to the same degree. A good rule 
that has been applied to initial installa- 
tions is to mount the projector as low 
as possible and the screen as high as pos- 
sible, consistent with sight lines to the 
screen from the rear of the orchestra, the 



mezzanine or balcony, and interference 
of the projector with these sight lines. 
This can be determined from the theater 
plans when the preliminary survey is 
made. 

When the distance from the balcony 
face to the screen is less than about 55 
ft the projector may have to be mounted 
back two or three rows in the balcony 



Stanko and Keen: Installing Theater TV 



347 




Fig. 4. Steelwork for mounting projector with legs in horizontal position, 

and service platform. 




Fig. 5. Projector with legs mounted vertically on balcony steel framework. 
348 March 1951 Journal of the SMPTE Vol. 56 




Fig. 6. Projector mounting with service platform, side view. 




Fig. 7. Projector mounting, front view. 
Stanko and Keen: Installing Theater TV 



349 



unless the screen can be moved far 
enough back on the stage to come within 
the operating range of the projector op- 
tical system. 

Mounting the projector back in the 
balcony will probably require some thea- 
ter constructional changes. 

Several ideas for mounting the pro- 
jector have been proposed for stadium- 
type theaters, but none have been tried 
in installations as of this writing. 

Installation of Projection Booth Rack 
Equipment (Fig. 9). The booth equipment 
consists of two racks 63 in. high which 
are fastened together with conduit nip- 
ples and interconnected electrically 
through the top conduit nipple. These 
are similar to sound equipment racks 
and have conduit knockouts at the top 
of each rack. 

It is desirable to mount the racks so 



that they are at a right angle to the 
front wall of the booth and adjacent to 
an observation port so that the screen 
can be observed when operating the 
controls. 

A voltage-regulator transformer 
mounted at some convenient location in 
the booth or adjacent to the booth, the 
main power switches, pilot lights and 
fuse box comprise the balance of the 
PT-100 System Booth Equipment. 

Booth space requirements in a few 
cases presented the only problem en- 
countered in the installation of the rack 
equipment. In these instances space 
was made available by rearrangement 
of booth equipment. 

Installation of the High-Voltage Sup- 
ply Unit. Some problems have been en- 
countered in the installation of the high- 
voltage supply unit. 




Fig. 8. Projector with horizontal leg mounting showing access to barrel 

from top. 



350 



March 1951 Journal of the SMPTE Vol. 56 



As yet there is no provision in the Na- 
tional Electrical Code covering the in- 
stallation of a unit of this type in thea- 
ters. Local rulings in several cities have 
required the unit to be placed in an en- 
closure separate from the projection 
booth with a door supplied with an elec- 
trical interlock switch which will turn 
off the supply when opened. Provision 
for this was anticipated in the equipment 
design. 

Generally, it is desirable to locate this 
unit as near the projector as practicable 
to keep the 80-kv cable as short as pos- 
sible. In some cases it has been lo- 
cated in a space under the balcony. The 
two high-voltage connectors (80-kv and 
20-kv) are made up at the time of in- 
stallation with a complete kit of materi- 
als furnished each installation. 




Fig. 9. Installation of Projection 
Booth Rack Equipment. 



A small chain hoist provided by the 
theater is installed above the high-volt- 
age supply unit to raise the assembly 
part way out of the oil for service access. 
The rectifier tubes and voltage-regulator 
tube are accessible from the top of the 
unit by raising a smaller part of the as- 
sembly making it unnecessary to raise 
the complete assembly to change tubes. 

Electrical Interconnection of Equip- 
ment Units. The exhibitor's installa- 
tion instructions furnished with the PT- 
100 include a complete wiring diagram 
with conduit sizes and all details re- 
quired by an electrician to wire the 
equipment (see Fig. 3). In many re- 
spects the wiring is similar to theater 
sound equipment and follows estab- 
lished theater practice. The wiring 
specifications conform to the National 
Electrical Code wherever it applies. 
This provision will generally meet all lo- 
cal regulations as well. 

Special instructions and assembly 
drawings are also given in the installa- 
tion instructions for making up the 
high-voltage 80-kv connectors and 20- 
kv connector. 

Installation of the Screen. At the pres- 
ent time perforated beaded screens are 
used for theater television projection, 
but the ultimate objective is to provide 
a screen which will satisfy the require- 
ments of theater television and motion 
picture projection with one screen. 
Such a screen is in prospect of realiza- 
tion. 

The screen must be located to provide 
a throw distance from the projector 
which will fall within the operating 
range of the PT-100 Optical System, as 
shown on the chart in Fig. 1. 

Since practical considerations may re- 
quire that the projector be located at 
the face of the balcony, it may be nec- 
essary to locate the theater television 
screen in front of or behind the present 
motion picture screen to bring the 
throw within the operating range of the 
system. 

This presents some problems and in a 



Stanko and Keen: Installing Theater TV 



351 



few cases has required the use of sepa- 
rate speakers for the theater television 
audio program. 

Another problem was posed by the 
requirement of tilting the screen an 
amount equal to the projection angle 
due to the short depth of focus in the ex- 
tremely "fast" optical system used in the 
PT-100 system. 

These problems have been resolved 
without much difficulty in installations 
to date. 

The screen must, of course, be masked 
to provide sharp picture edges. The 
angle of projection and the tilt of the 
screen and its location are all calculated 
in advance of the installation from the 
data obtained in the preliminary survey. 

Installation Check and Operation 

A very definite procedure has been 
established to check and adjust the 
equipment in an orderly step-by-step 
routine when the installation is com- 
pleted and power is ready to be turned 
on. 

This is done by a field engineer who 
has received previous training and is 
provided with the required equipment 
to perform this operation. 

One of the important checks made on 
the equipment prior to installing the 
7NP4 Kinescope is a step-by-step pro- 
cedure to be sure that the kinescope 
protection circuits are functioning cor- 
rectly. 

When the equipment has been thor- 
oughly checked and all adjustments 
made, the projectionists are given in- 
structions in operation techniques. 

Proper operation of the equipment is 
not difficult and in many respects par- 
allels operation of a home television re- 
ceiver. Projectionists have done an 
excellent job on this in the few locations 
involved so far. 

Recently, a group of projectionists 
selected by the I.A.T.S.E. from practi- 
cally every state in the Union were given 
an intensive course of television instruc- 
tion in Camden, and all had an oppor- 



tunity to operate the PT-100 equip- 
ment set up especially for the purpose 
in the RCA Engineering Laboratory 
Theatre. 
Service and Maintenance 

In any electronic or mechanical 
equipment there is always the problem 
of maintaining performance at certain 
predetermined standards. 

Theater television equipment, due to 
its complexity in comparison with 
sound motion picture equipment, re- 
quires more attention in several respects 
to insure optimum performance. 

There are over 100 tubes in the PT- 
100 Theater TV System. The average 
sound motion picture equipment has 
only about 20 tubes. 

While the PT-100 equipment has 
been carefully designed for stability of 
operation consistently over long periods, 
some service operations must be per- 
formed by skilled engineers at periodic 
intervals to insure consistently top per- 
formance. 

High voltages applied to the 7NP4 
Kinescope could cause serious damage to 
the tube if the protection circuits did not 
function properly. These tubes are 
comparatively expensive, therefore pe- 
riodic checks must be made in the pro- 
tection circuits with proper test equip- 
ment to be sure that the correct circuit 
constants are always maintained. Pulse 
forms and amplitudes must be checked 
periodically with an oscilloscope suitable 
for this work and adjustments made if 
required. 

The projector mirror and correction 
lens X-ray plate combination must be 
cleaned periodically. This requires 
careful technique with cleaning solutions 
developed especially for these surfaces. 
For example, the cleaning solution used 
on the mirror cannot be used on the lens 
surface. Service procedures are being 
formulated and will soon become rou- 
tine. 

Close coordination between the RCA 
Theater Equipment Engineering Group, 
responsible for the development and de- 



352 



March 1951 Journal of the SMPTE Vol. 56 



sign of the PT-100 Equipment and the 
Technical Products Service Group of the 
RCA Service Company, resulted in in- 
corporating many features in the equip- 
ment to make installation problems less 
difficult and expensive, and to facilitate 
service and maintenance operations. 

Test Equipment 

While many troubles in television 
equipment can be diagnosed by symp- 
toms appearing on the picture screen, it 
is generally necessary to use an oscil- 
loscope to trace them down in the sus- 
pected circuits. 

It is necessary to use an oscilloscope 
that will give a true picture of the wave 
forms in any part of the circuits. The 
Type-58 or -79 Oscilloscope is used by 
the RCA Field Engineers for this pur- 
pose. Some of the less expensive 
"scopes" which are suitable for home 
television service work are not suitable 
for theater television equipment. Usu- 
ally these less expensive instruments are 
designed for high sensitivity in the ver- 
tical amplifier at the expense of wide 
band response and they have other de- 
ficiencies which rule out their use for 
theater television equipment although 
they may be entirely satisfactory for 
checking home television receivers. 

A high-resistance voltmeter is also a 
requirement. The 165A VoltOhmyst is 
used for this purpose. An instrument 
of this type has been found to be essen- 
tial in checking this equipment. 

A portable synchronizing and grating 
generator for field use is now under de- 
velopment. This instrument will fur- 
nish a suitable video signal to test the 
PT-100 Equipment when other signals 
are not available. Other test equip- 
ment is made available when required. 

Discussion 

HENRY ROGER: Did Mr. Stanko say 
it would be simpler to use rear projection 
on a translucent screen? I presume that 
you may get higher field intensity. 

OTTO H. SCHADE: I think you are per- 
fectly correct in assuming that that is 



Bible. The normal light gain of a 
beaded screen is not very high in the 
order of 2 or 2^. With a properly de- 
signed rear-projection screen of the lens 
type, you can get gains as high as 15 if 
the viewing angle from the front isn't 
too wide. Such screens are very expen- 
sive. They have to be made very precise 
and considerable space must be provided 
behind the screen for the projection dis- 
tance. If the theater were laid out for 
such projection requirements, then it 
might pay to look into this situation. I 
think there are some rather early papers 
in the art. I remember a German paper 
reporting actual gains of the order of 12 
that is, the intensity was 12 times that 
obtained from a purely diffuse reflecting 
screen. 

MR. ROGER: I believe there is one dis- 
advantage with translucent screens, and 
that is if the projection is in line with the 
eye of the observer, you get a hot point; 
but I believe that this can be overcome 
by mounting the projector slightly lower. 

MR. SHADE: The type of screen which 
I mentioned as a lens screen avoids that 
completely. It is a structure of small 
lenses similar to a Fresnel lens directing 
all the rays from the sides as well as the 
center of the screen into a certain angle 
where the seats are. You can design it 
so that the intensity distribution in this 
viewing field is completely uniform. A 
normal translucent screen has no such 
means. It acts, for example, like a 
frosted glass, which has a very definite 
lobe (like an antenna) producing a hot 
spot in the center and much lower in- 
tensity at the sides. By putting a large 
condenser lens behind it, you can direct 
the side rays toward the center and there- 
fore create a small area where the screen 
brightness is uniform. A lens screen is 
designed to widen this angle over a wider 
area without having the loss due to frost- 
ing the screen which absorbs light and 
scatters it vertically as much as horizon- 
tally. The lens system can be made to 
keep the vertical scattering extremely 
small and thus concentrate the entire 
light energy into the desired viewing 
space and in that manner raise the ef- 
fective brightness. Lens screens can also 
be designed for front projection, but as I 
said, a lens screen is very expensive com- 
pared to normal low-gain screens. 



Stanko and Keen: Installing Theater TV 



353 



Standards 



Splices for 16-Mm and 8-Mm Film 



THESE PROPOSED American Standards, 
developed by the 16-Mm and 8-Mm 
Motion Pictures Committee, appear on 
the following pages. They are pub- 
lished here for trial and criticism for a 
period of ninety days. Please forward 
any comments to Henry Kogel, Staff 
Engineer at Society Headquarters, by 
July 1, 1951. 

For many years the standards for 
splices have shown both a diagonal 
splice with a 0.070-in. overlap and a 
straight splice with a 0.100-in. overlap. 
It is presumed that the diagonal splice 
was narrower so that it would not en- 
croach on the perforations. These di- 
mentions were specified in Chart 12 of 
Z22-1930 and repeated in 1941 in 
Z22.24 and Z22.25. Actually, however, 
the dimensions of the straight splice 
were first established in 1924 when 
16-mm equipment was introduced. 
Many thousands of splicers making 
splices with these dimensions have been 
manufactured and used. When splicers 
for 8-mm film were introduced in 1932, 
the same 0.070-in. and 0.100-in. over- 
laps for diagonal and straight splices 
were specified. 

For twenty years these dimensions 
have been retained as new splicers came 
on the market, even though a great deal 
of experimental work was being con- 
ducted in an effort to improve splices 
and lengthen their life during projection. 

In January, 1944, Subcommittee C of 
the American Standards Association's 



Z52 War Committee on Photography 
and Cinematography was appointed to 
handle standards relating to 16-mm lab- 
oratory practice. One of the first proj- 
ects of this subcommittee was to 
specify the type of splice that should be 
used for 16-mm release prints. The 
discussion showed that there was little 
agreement among the laboratories. Some 
preferred narrower splices because they 
are less conspicuous on the screen. It 
was stated that, contrary to former be- 
liefs, wide splices were more susceptible 
to failure than narrow ones. Several 
members said that they were using 
splices that were less than 70-mils wide 
and had found them quite satisfactory. 
It was then agreed to make the 0.070-in. 
overlap a maximum inasmuch as it was 
desired to secure splices with as small an 
overlap as possible. It was then decided 
to write a new standard covering diag- 
onal and straight splices. 

Later the title of the proposed stand- 
ard was changed to limit the splices to 
processed film because the suitability of 
the splices for raw stock had not been 
considered by the committee. The final 
War Standard, Z52.20-1944, was ap- 
proved May 29, 1944, by the American 
Standards Association. 

This action resulted in the existence 
of two standards for straight splices : the 
original 0. 100-in. splice from Z22.24-1941 
and Z22.25-1941, and the new 0.070-in. 
splice in Z52.20-1944. It was realized 
at the time that careful comparative 



354 



March 1951 Journal of the SMPTE Vol. 56 



tests would eventually be required in 
order to determine which splice was 
superior. If it were decided to carry 
over into peacetime the 0.070-in. splice, 
it would be necessary to rebuild all the 
existing tools for amateur splicers, 
which would involve considerable ex- 
pense to the industry because of the 
many splicers for 0.100-in. splices on the 
market. 



Consideration of Dual Standards 

The desirability of continuing the two 
standards for splices was considered 
when Committee Z22 of the American 
Standards Association reviewed the War 
Standards in October, 1945. At this 
meeting, and at subsequent meetings, 
the need for having standards for splices 
was questioned, but the committee de- 
cided that such standards were of value 
to the industry. Eventually, the ques- 
tion was referred by Z22 to the Stand- 
ards Committee of the Society, and a 
subcommittee was appointed, under the 
chairmanship of W. H. Offenhauser, 
Jr., to review the situation. 

Meanwhile, the suggestion had been 
made that the 0.100-in. splice be desig- 
nated the amateur splice and that the 
0.070-in. be the professional or labora- 
tory standard. This suggestion was dis- 
cussed by the subcommittee, and it was 
agreed that the standard should pre- 
scribe good commercial practice, as well 
as a splice that would be practicable for 
the amateur. 

The relative merits of symmetrical 
versus unsymmetrical splices were also 
discussed, but it was decided that both 
diagonal and straight splices would be 
shown symmetrical with respect to the 
included perforation. Finally, although 
it was brought out that the 0.070-in. 
splice would be difficult for the amateur 
to perform because of the more precise 
control of hand scraping required, it was 
voted that the 0.070-in. straight splice 
be standardized for a trial of one year. 
During that period, data would be col- 



lected so that a review could be made and 
a final decision reached. 

This decision of the subcommittee was 
reported to the Standards Committee at 
a meeting on February 20, 1946. Al- 
though one member reported at that 
time that the narrower splice did not 
last as long in projection as the 0.100-in. 
splice, it was decided to publish the 
standard calling for the narrower splice 
and to collect more information during 
the trial period of one year. 

When the report on splices was pre- 
sented to the Society on May 8, 1946, 
the following test results was submitted 
to Mr. Offenhauser: Of 45 loops of film, 
each containing one 0.100-in. and one 
0.070-in. splice and run in a projector 
until one of the splices broke, 40 loops 
broke first at the 0.070-in. splice and 
only 5 broke first at the 0.100-in. splice. 
These tests were run by two different de- 
partments of the same company, and in- 
cluded three projectors and splices made 
by several different operators. A third 
group ran 6 loops containing two splices 
of each width. The average life of the 
0.070-in. splice was half that of the 
0.100-in. splice. 

Conclusions of Tests 

These conclusions were confirmed 
later with splices made on an expensive 
precision splicer of the 0.070-in. type. 
Seven loops were made with one 0.070- 
in. precision splice and one 0.100-in. 
splice. All broke at the 0.070-in. splice 
first. In most cases, nevertheless, the 
life of the weaker 0.070-in. splice was 
probably adequate for either amateur or 
professional use. Such might not have 
been the case, however, if an ordinary 
hand splicer had been employed since 
most operators have considerable diffi- 
culty scraping the narrow splice. Be- 
because the strip of film between the 
perforations and the cut end of the film 
is only 0.010-in. wide, it is difficult to re- 
move all the emulsion without damag- 
ing the film. In addition, shrinkage of 



Standards: Splices for 16-Mm and 8-Mm Film 



355 



the film reduces this width still further 
and makes the scraping operation even 
more difficult. 

From the foregoing tests, it is quite 
apparent that the wide splice was 
stronger despite its greater resistance to 
smooth bending of film around loops, 
rollers and sprockets. 

Comments After Publication 

The proposal of the subcommittee 
was published in the July, 1946, JOUR- 
NAL. It was simpler than previous 
standards because nonessential dimen- 
sions were omitted. Although only 
sound film was shown, it was stated that 
splices for silent film would be the same, 
A curved splice, included in the War 
Standard, was dropped because no one 
expressed interest in it. Comments were 
invited, particularly on the relative de- 
sirability of the 0.070-in. splice com- 
pared with the 0.100-in. splice, and on 
symmetrical versus unsymmetrical 
splices. 

Although the response was rather 
small, a number of interesting sugges- 
tions were offered. The suggestions fell 
into two distinct groups: first, those 
concerned with splices made in process- 
ing laboratories by professionals, usu- 
ally in negative or other preprint ma- 
terial; and second, those concerned 
with splices made in reversal originals or 
release prints. 

Of the first group, a proposal by 
Joseph V. Noble of DeFrenes & Com- 
pany Studios called for an invisible 
0.016-in. frameline splice in the width of 
the film devoted to the picture area, 
with an overlap of 0.079 in. along both 
edges of the films. In addition, John S. 
Carroll listed the advantages of the un- 
symmetrical Griswold "negative" splice, 
and advocated increasing its width from 
0.0625 in. to 0.070 in. for greater 
strength. He did not recommend it for 
release prints, however, because of its 
poorer mechanical properties. G. A. 
Chambers recalled the successful use of 



0.070-in. splices at the Anacostia labora- 
tory and recommended them for labora- 
tories but not for the amateur. 

The group favoring the retention of 
the 0.100-in. overlap was interested pri- 
marily in amateur splicing. Their argu- 
ments were based on the longer projec- 
tion life, the greater success that the 
amateur has in making the 0.100-in. 
splice, and the large investments in tools 
for 0.100-in. splicers. 

Since the industry did not appear to 
be ready for the adoption of the stand- 
ards outlined in the July, 1946, JOUR- 
NAL, the Committee on Standards 
agreed that the original Z22.24 and 
Z22.25 should be reaffirmed until the 
situation is clarified. At this point, the 
problem of standards for splices was re- 
ferred to the 16-Mm and 8-Mm Motion 
Pictures Committee. 

Work of 16-Mm and 8-Mm Committee 

In preparing the drafts submitted 
herewith, the Committee noted that the 
old standards did not conform to mod- 
ern methods of dimensioning. It was 
suggested, also, that sound and silent 
films be combined into one standard. As 
these changes were made, it was dis- 
covered that other alterations and addi- 
tions would be desirable. For example, 
the splices were redrawn so that they 
would appear as they do on splicing 
blocks. Tolerances were added to cover 
the proper transverse alignment of the 
perforations and the edges of the films, 
and the parallelism of the edges of the 
two films after the splice has been made. 
Again the question of whether straight 
splices should be 0.070-in. or 0.100-in. 
wide was raised. For 8-mm film the 
0.100-in. splice was of course preferable. 
When the title and wording of the 16- 
mm splice proposal were changed to 
limit it to release prints, all objections to 
the 0.100-in. splice were withdrawn. 

Drafts of the proposals approved by 
the 16-Mm and 8-Mm Motion Pictures 
Committee were submitted to the Stand- 



356 



March 1951 Journal of the SMPTE Vol. 56 



ards Committee in May, 1949, for bal- 
loting on the question of publication in 
the JOURNAL for a ninety-day period of 
trial and comment. In due course, this 
ballot was completed favoring publica- 
tion at an early date. During the course 
of this balloting, however, various mem- 
bers of the 16- and 8-mm Committee 
raised the question of whether or not 
manufacturers of splicers still promoted 
the sale of equipment to make diagonal 
splices. Upon investigation, it was 
found that no manufacturer expects to 
produce equipment of this type in the 



future. In the light of this information, 
it was recommended that the diagonal 
splice be dropped from the proposal. 

Since this was considered to be a 
major change from the version upon 
which the Standards Committee origi- 
nally balloted, the ballot was withdrawn 
at a meeting of the Standards Commit- 
tee on February 1, 1950. In subsequent 
balloting, completed in December, 1950, 
and January, 1951, the Committee ap- 
proved publication of these two pro- 
posals in their present form for trial and 
comment. 



Erratum: 

W. F. Kelley and W. V. Wolfe, "Recent studies on standardizing the 
Dubray-Howell Perforation for universal application," Jour. SMPTE 
vol. 56, pp. 30-38, Jan. 1951. 

Page 32, line 5 et seq., and Fig. 1 : For Cooke read Cook. This refers to 
Allen W. Cook who made the Ansco proposal a number of years ago. 



Standards: Splices for 16-Mm and 8 -Mm Film 



357 



Proposed American Standard 

Splices for 
16-Mm Motion Picture Films 

for Projection 



PH22.24 

(227.24) 

Revision of 

Z22.24 - 194! 



Scope. Splices made in accordance with this 
standard are primarily for use with films in- 
tended for actual projection, such as release 
prints and reversal films. It is not intended 
that this standard be prejudicial to the use of 



P. 1 of 2 ff . 

diagonal type splicers, nor to the use of nar- 
rower splices for professional purposes. For 
negatives and other laboratory films, nar- 
rower splices, sometimes with one edge on the 
frameline, frequently are used. 

Silent film has perforations 
along this edge also. "V 




Inches 


Millimeters 


A 0.100 ^0005 


2 - 54 -SS 


B 0.548 
0.001 


>rt + 0.025 
13 ' 920 - 0.025 


C OVA +-0 
-0.003 


ftr , +0.00 
-0.08 


D . 324 +0.000 


823 +0.00 
-0.08 



NOT APPROVED 



358 



March 1951 Journal of the SMPTE Vol. 56 



Proposed American Standard 

Splices for 

16-Mm Motion Picture Films 



for Projection 



PH22.24 

(Z22.24) 

Revision of 

Z22.24 - 1941 

and 
Z22.25 - 1941 



Note 1. In the plan view, the splice is arranged with 
the perforations at the bottom in order to show them 
as they appear on most splicers. The splice may be 
made with the films turned through an angle of 180 
degrees, or any other angle, but of course the emul- 
sion surface should always be up. It is customary to 
scrape the top (emulsion) surface of the left-hand film 
and to cement this scraped area to the bottom (base) 
surface of the right-hand film. 

Note 2. Dimension A is given a negative but no 
positive tolerance because narrower splices are less 
conspicuous on the screen and are less likely to affect 
the normal curvature of the film as it follows the 
bends in its path through cine-machinery. 

Note 3. Dimension B controls the longitudinal regis- 
tration of the two films being spliced. It is measured 
to the perforations that are most commonly used for 
registration on splicing blocks, and to the nearer 
edges of these perforations because they are edges 
that are generally used for the registration. This di- 
mension is made the same as in 222.77, Splices for 
8-Mm Motion Picture Film, because many splicers are 
designed to accept either 16- or 8-mm film. 

The nominal value of the B dimension was made 
0.548 inch instead of the usual 0.550 (for unshrunk 
film) because the films being spliced are always 
shrunk to some extent. The 0.548 figure corresponds 
to a shrinkage of 0.36 percent, while the 0.549 and 
0.547 values, permitted by the tolerances, correspond 
to 0.18 percent and 0,55 percent, respectively. Thus, 
the tolerances include the range of shrinkage ordi- 
narily encountered when film is being spliced. 



P. * f S p 

Note 4. Dimensions C and D were chosen to give a 
straight 0.100-inch splice that is symmetrical about 
the included perforation (and, therefore, the frame- 
line) when the film is shrunk 0.36 percent. See Note 
3 above. 

Note 5. The width of the film at the splice shall not 
exceed 0.630 inch. If the film has been widened dur- 
ing scraping, the extra width shall be removed. 

Note 6. The overlapping perforations of the two 
films shall not be offset laterally more than 0.002 
inch. 

Note 7. At the splice, the edges of the two spliced 
films shall not be offset laterally more than 0.002 
inch, unless a difference in the lateral shrinkages of 
the two strips makes it impossible to maintain that 
tolerance. Shoulders formed by such misalignment 
shall be beveled after the cement has dried. 

Note 8. In the plan view, the angle between the 
respective edges of the spliced films shall be 180 de- 
grees, plus or minus 40 minutes. Thus, the spliced film 
shall be aligned to the extent that when one portion 
of the film is placed against a straight edge, the 
other portion will not deviate more than 0.006 inch 
(approximately the thickness of the film) in 6 inches. 

Note 9. In order to prevent the appearance of a 
white line on the screen, the scraped area shall be 
0.001 to 0.003 inch narrower than the area covered 
by the overlapping film. The presence of this narrow 
uncemented area will not shorten the life of the splice. 



NOT APPROVED 



March 1951 Journal of the SMPTE Vol. 56 



359 



Proposed American Standard 

Splices for 
8-Mm Motion Picture Films 



PH22.77 

(Z22.77) 



P. 1 to pp. 



HM 



/DODO 









Inches 


Millimeters 


A 


+ 0.000 
- 100 -0.005 


'* t ^ 


B 


+0.001 
' 548 -0.001 


13.920 1 0.025 


C 


. +0.000 
' 324 -0.003 


8 ' 23 -SS 


D 


t +0.000 
' 324 -0.003 


8 ' 23 -SS 



Note 1. In the plan view, the splice is arranged with 
the perforations at the bottom in order to show them 
as they appear on most splicers. The splice may be 
made with the film turned through an angle of 180 
degrees or any other angle, but, of course, the emul- 
sion surfaces should always be up. It is customary to 
scrape the top (emulsion) surface of the left-hand 



film, and to cement this scraped area to the bottom 
(base) surface of the right-hand film. 

Nate 2. Dimension A is given a negative, but no 
positive, tolerance because narrower splices are less 
conspicuous on the screen and are less likely to affect 
the normal curvature of the film as it follows the 
bends in its path through cine-machinery. 



NOT APPROVED 



360 



March 1951 Journal of the SMPTE Vol.56 



Proposed American Standard 

Splices for 
8-Mm Motion Picture Films 



PH22.77 

{Z22.77) 



Note 3. Dimension B controls the longitudinal regis- 
tration of the two films being spliced. It is measured 
to the perforations that are most commonly used for 
registration on splicing blocks, and to the nearer 
edges of these perforations because they are the 
edges that are generally used for the registration. 
This dimension was made the same as in 222.24, 
Splices for 16-Mm Motion Picture Film, because many 
splicers are designed to accept either 8-mm or 16-mm 
film. 

The nominal value of the B dimension was made 
0.548 inch instead of the usual 0.550 (for unshrunk 
film) because the films being spliced are always 
shrunk to some extent. The 0.548 figure corresponds 
to a shrinkage of 0.36 percent, while the 0.549 and 
0.547 values, permitted by the tolerances, correspond 
to 0.18 percent and 0.55 percent, respectively. Thus 
the tolerances include the range of shrinkage ordi- 
narily encountered when film is being spliced. 

Note 4. Dimensions C and D were chosen to give a 
0.100-inch splice that is symmetrical about the in- 
cluded perforation (and therefore the frameline) when 
the film is shrunk 0.36 per cent. See Note 3. 

Note 5. The width of the film at the splice shall not 



P. 2 of 2 pp. 

exceed 0.317 inch. If the film has been widened dur- 
ing scraping, the extra width shall be removed. 

Note 6. The overlapping perforations of the two 
films shall not be offset laterally more than 0.002 
inch. 

Note 7. At the splice, the edges of the two spliced 
films shall not be offset laterally more than 0.002 inch 
unless a difference in the lateral shrinkages of the 
two strips makes it impossible to maintain that toler- 
ance. Shoulders formed by misalignment shall be re- 
moved after the cement has dried. 

Note 8. In the plan view, the angle between the re- 
spective edges of the spliced films shall be 180 de- 
grees, plus or minus 40 minutes. Thus, the spliced film 
shall be aligned to the extent that when one portion 
of the film is placed against a straight edge, the other 
portion will not deviate more than 0.006 inch (ap- 
proximately the thickness of the film) in six inches. 

Note 9. In order to prevent the appearance of a 
white line on the screen, the scraped area shall be 
0.001 to 0.003 inch narrower than the area covered 
by the overlapping film. The presence of this narrow 
uncemented area will not shorten the life of the splice. 



NOT APPROVED 



March 1951 Journal of the SMPTE Vol. 56 



361 



69th Semiannual Convention 



AFTER THE ADVANCE NOTICE of the Convention, listing ten technical sessions, 
went out to Society members on March 9, two sessions were added to make 
the Tentative Program which has also been mailed. The Tentative Program 
contained 56 papers and 7 committee reports. Additional copies of the Tenta- 
tive Program are available from Society headquarters. 

Adjustments, indicated as likely at JOURNAL press time, have been taken into ac- 
count to show below the transfer of the Screen- Viewing Factors and the Film Projec- 
tion Symposiums to Wednesday, concurrent with Wednesday's High-Speed Photog- 
raphy Sessions, and to show that Thursday afternoon will probably be used to take up 
some of the papers shown in the heavy sessions in the Tentative Program. 



APRIL 30 MAY 4 

Monday afternoon 
Monday evening 
Tuesday morning 
Tuesday afternoon 

Tuesday evening 
Wednesday morning 

u 

Wednesday afternoon 

U li 

Thursday morning 
Thursday afternoon 
Friday morning 
Friday afternoon 



Film and Processing 

Motion Picture Techniques 

Television Recording and Reproduction 

Television Session and Tour of the Bell Telephone 

Murray Hill Laboratories 
Television and Motion Picture Production 
High-Speed Photography 
Screen- Vie wing Factors Symposium 
High-Speed Photography 
Film Projection Symposium 
High-Speed Photography 
See Final Program 
Magnetic Recording 
Sound Recording 



Each session will open with a motion 
picture short. A luncheon will open the 
Convention on Monday noon, and a 
banquet and dance will be held Wednesday 
evening. Complimentary tickets to se- 
lected Broadway motion picture theaters 
will be issued to registered members and 
guests. 

Any reader who did not receive the 
Advance Notice can get a copy from 
Society headquarters or he can write for 
reservations (mentioning the SMPTE) 
direct to: Front Office Manager, Hotel 
Statler, 7th Ave., 32d and 33d Sts., 
New York 1. 



Also, readers of subscription copies of 
the JOURNAL need only send a postal to 
Society headquarters to get a copy of the 
Tentative Program. 

SPADEWORK 

Bill Kunzmann, Convention Vice-Presi- 
dent, notes that the Papers Committee, 
listed in the JOURNAL last month, has 
turned in a fine job. He urges that all 
who have been asked to help with work 
assigned to the following chairmen dig in 
with a will. With their help, all Conven- 
tion machinery will function smoothly, 
thanks to these chairmen: 



Hotel and Transportation H. D. Bradbury 
Local Arrangements E. M. Stifle 
Luncheon and Banquet W. B. Lodge 
Membership and Subscriptions L. E. Jones 
Motion Pictures Emerson Yorke 



362 



Public Address H. B. Braun 

Publicity Harold Desfor with Leonard Bid well and Harry Sherman 

Projection, 35-mm H. E. Heidegger, with officers and members of New York Pro- 
jectionists Local 306 

Projection, 16-mm T. P. Dewhirst 

Registration and Information E. R. Geib with P. D. Reis, G. H. Gordon and E. A. 
Hungerford 

Television C. L. Townsend 

Ladies' Registration Mrs. E. M. Stifle, Hostess, with Mrs. E. I. Sponable, Mrs. 
Herbert Barnett and Mrs. O. F. Neu 



Engineering Activities 



STANDARDS COMMITTEE MEETING 

At its annual meeting held in the last 
week of January, 1951, the Standards 
Committee, under the Chairmanship of 
Frank Carlson, had a very full and fruitful 
session. The first order of business was a 
review of the status of all Proposed 
Standards currently before the Committee, 
as outlined below: 

Approved by Standards Committee and 
PH22 

Cutting Dimensions for 32-Mm on 35-Mm 
Motion Picture Negative Raw Stock, 
PH22.73 

Zero Point for Focusing Scales on 16-Mm 
and 8-Mm Motion Picture Cameras, 
PH22.74 

Mounting Threads and Flange Focal Dis- 
tances for Lenses for 16-Mm and 8-Mm 
Motion Picture Cameras, PH22.76 

Approved by Standards Committee for Sub- 
mittal to PH22 

Sound Transmission of Theater Projection 
Screens, PH22.82 

Approved by Standards Committee for Pre- 
liminary Publication in JOURNAL 

A and B Windings of 16-Mm Raw Stock 
Film, PH22.75; published January, 1951 

Edge Numbering of 16-Mm Motion Pic- 
ture Film, PH22.83; published January, 
1951 

Splices for 16-Mm Films for Projection, 
PH22.24; published in this JOURNAL 

16-Mm Motion Picture Projection Reels, 
PH22.11; published February, 1951 

Dimensions for Projection Lenses, Medium 
Prefocus Ring Double-Contact Base-Up 



Type, PH22.84; published February, 
1951 

Dimensions for Projection Lamps, Me- 
dium Prefocus Base-Down Type, 
PH22.85; published, February, 1951 

Splices for 8-Mm Motion Picture Film, 
PH22.77; approved subsequent to meet- 
ing and published in this JOURNAL. 

Edge Guiding of 16-Mm Films. The 
question of edge guiding was then explored 
in view of an apparent inconsistency in 
Society policy. In this regard recent 
Standards, PH22.7-1950, and PH22.8- 
1950, do not specify a single guided edge, 
stating that either edge may be used and 
giving the arguments for each; however, 
other Standards, PH22.15, PH22.16 and 
PH22.41, do specify the guided edge. The 
first two deal entirely with emulsion posi- 
tion and hence the Committee voted to 
revise these standards, deleting specifica- 
tion of guided edge. However, PH22.41, 
Sound Records and Scanning Area of 16- 
Mm Sound Motion Picture Prints, pre- 
sents several problems: 

1. The present tolerances may be de- 
pendent on the specification of the guided 
edge. 

2. Advent of 32-mm film as source of 
16-mm release prints and consequent slit- 
ting problems may require such specifica- 
tion. It was, therefore, agreed that the 
Engineering Vice-President would refer 
this to the Sound Committee (to be re- 
viewed by the 16-Mm and 8-Mm Com- 
mittee) for further study with the directive 
to revise if this can be done without de- 
grading the present standard. 

Glossary. The desirability of compiling 



363 



a glossary of technical terms peculiar to 
the motion picture industry was readily 
agreed upon. The discussion centered 
rather on practical methods of achieving 
this, inasmuch as attempts had been made 
in the past without appreciable success. 
It was finally agreed that each committee 
chairman would be asked to draw up a 
list of terms he considered vital for inclu- 
sion in such a glossary. This could be 
readily achieved and would provide a basis 
for further work. It might then be possi- 
ble, after some discussion, to get imme- 
diate agreement on a definition for 75-90% 
of the terms. A glossary would then exist 
and, although incomplete, be at least 
better than nothing at all and a good 
starting point for further work. 

International Standards. Fred Bow- 
ditch, Engineering Vice-President, gave a 
brief history of the International Stand- 
ards .Organization (ISO), noting that the 
International Standards Association dis- 
solved at the onset of the second World 
War and that the ISO was formed as an 
adjunct to the UN. The American Stand- 
ards Association (ASA) is the United 
States representative in the ISO and as 
such holds the Secretariat of Technical 
Committee 36 (TC36) for Motion Pictures. 
This means that sectional committee 
PH22 of ASA should rightfully be han- 
dling ISO questions, but in effect, as spon- 
sor of PH22, the final responsibility belongs 
to SMPTE. A meeting of TC36 had been 
proposed for this coming summer in 
Geneva, but the consensus was that in the 
absence of a specific agenda of recognized 
importance, there was no real justification 
for such a meeting now. As Secretariat, 
ASA had distributed 40 American Stand- 
ards in 1948 and proposed they be made 
International Standards. Comments on 
these have been received from several 
countries and require additional corre- 
spondence on our part. In view of pro- 
jected ISO meetings in this country during 
1952, it might be wise to begin now the 
preparation of specific agenda. 

The ensuing discussion brought out 
several vital points: 

1. Some of the standards submitted 
have since been revised. 

2. The advent of low-shrink film may 
soon require revision of other standards. 

3. Additional standards now exist which 



may also warrant international standardi- 
zation. It was therefore agreed that all 
members of the Standards Committee 
would inform the Chairmen of those 
standards which they feel might be im- 
portant for an International Standards 
Program and include along with the 
standard a brief review of the history and 
present status of each standard. 

Very often, material is received from 
other nations with requests for comments. 
In such a case, it was agreed that the 
proper channels for such material should 
be from the various foreign standards 
institutes to ASA (PH22) then to the 
SMPTE Engineering Vice-President and 
finally to the appropriate engineering 
committee. 

Research Projects. The Engineering 
Vice-President stated that he had been 
requested by the Board of Governors, in 
line with a recommendation by Clyde 
Keith, to prepare a list of research prob- 
lems which might be used by graduate 
students in universities as theses projects. 
Such graduate research activity can well 
produce information of value to industry 
and is thus worth encouraging. He there- 
fore emphasized his earlier written request 
that each engineering committee consider 
the preparation of a list of projects in its 
field, together with brief background 
material explaining each project. 



NEW STANDARDS SOON 

Three proposed American Standards are 
now on their last lap hi the course of be- 
coming Approved Standards. Having 
been approved by the Engineering Com- 
mittee, the Standards Committee, ASA 
Sectional Committee PH22 and the Soci- 
ety Board of Governors, they are now 
awaiting final approval by the ASA Photo- 
graphic Standards (Correlating) Commit- 
tee and then the ASA Standards Council. 
They are: 

1. PH22.73, Cutting and Perforating 
Dimensions for 32-Mm on 35-Mm Raw 
Stock 

2. PH22.74, Zero Point for Focusing 
Scales for 16-Mm and 8-Mm Motion 
Picture Camera 

3. PH22.76, Mounting Threads and 
Flange Focal Distances for Lenses for 
16-Mm and 8-Mm Motion Picture Cameras 



364 



BOOK REVIEW 



Acoustical Designing in 
Architecture 

By Vern 0. Knudsen and Cyril M. Harris. 
Published (1950) by John Wiley, 440 
Fourth Ave., New York 16. 404 pp. + 
45 pp. appendix + 7 pp. index. 180 illus. 
+ 8 tables. 5% X 8& in. Price $7.50. 

This is the most readable, useful and 
practical book on architectural acoustics 
which we have encountered. Naturally, 
most of the information can be found in 
more complete and complicated form in 
the Acoustical Society Journals, earlier 
books and other periodical literature. 
Such a collection, we believe, would be far 
too extensive and technical for the average 
architect and engineer (other than the 
acoustical engineer) to maintain or use to 
his advantage. 

Professor Knudsen's earlier book, Archi- 
tectural Acoustics, embracing much of his 
original research, furnishes fit foundation 
for this new and excellent collaboration. 
Comparison of the two in such matters as 
sound absorptive materials shows that 
many of the materials mentioned in the 
first book have disappeared from the 
market, while new ones (they generally 
have coined names) have been developed 
by a fast-moving industry even since the 
publication of the present book. Changes 
such as this can be expected in materials, 
methods and sound systems, but certainly, 
the well-presented basic information will 
remain useful. 

The authors have designed this book pri- 
marily for architects and students of archi- 
tecture, and there is no doubt that if an 
architect will make use of this book in- 
telligently, he will avoid most of the glar- 
ing errors which crop up in public build- 
ings. We suggest that Chapters 9 through 
14 be read before consigning the book to a 
shelf. This procedure will probably arouse 
enough interest to make the reader do the 
first eight chapters, at least lightly. 
Chapters 1 through 8 deal largely with the 
physics of sound and with the nature of 
speech and hearing. Chapters 9 through 



14 are concerned with basic design, the 
selection of a proper site, arrangement of 
rooms, control of noise both air-borne and 
structurally transmitted, and use of sound 
amplification systems. These parts to- 
gether offer a fairly simple and lucid 
explanation of the nature and behavior of 
sound in its relation to architecture. 
Specific problems in rooms for special uses 
are treated in the remainder of the book. 
Reference may be made to these sections 
as the need requires. 

There is a chapter on each of the follow- 
ing: auditoriums; school buildings; com- 
mercial and public buildings; homes, 
apartments and hotels; church buildings; 
broadcasting, television and sound-record- 
ing studios. 

While this book purports to be a non- 
mathematical treatise, it frequently steps 
out of character with the sudden appear- 
ance of mathematical terms not commonly 
met with in architecture. These consti- 
tute the common language of the acousti- 
cal engineer or physicist but probably 
convey little information to the architect. 
However, many of the expressions given 
are interpreted by graphs which, if studied, 
will supply a fairly close answer. 

Undoubtedly, complicated problems will 
arise where the architect may need the 
services of a specialist in this field, but his 
task will be considerably lightened and his 
resulting design better if he has a basic 
understanding of the acoustic principles 
underlying the form and structure being 
considered. 

Aside from the use of this book by the 
architect and reader of general interest, it 
is recommended as a good additional refer- 
ence work for engineers in the sound field. 
The careful listing of design procedure, the 
extensive collation of sound-absorbing 
coefficients of different materials and data 
on sound transmission of various construc- 
tions make the book very much worth 
while. JAMES Y. DUNBAR, William J. 
Scully Acoustics Corp., 101 Park Ave., 
New York 17. 



SMPTE Officers and Committees: New rosters are scheduled to be published 
in the April JOURNAL. 



365 




Virgil B. Sease 



AFTER 33 YEARS with E. I. du Pont de 
Nemours & Company, Dr. Sease retired 
from his activities at Du Font's Redpath 
Laboratory, Parlin, N.J., at the end of 
1950. He has been a member of this 
Society since 1926. He became a Fellow 
in 1934, the year in which the grade of 
Fellow was established. 

Dr. Sease joined the Du Pont Company 
in 1917 as a research chemist at the Ex- 



perimental Station near Wilmington, Del. 
From late in 1917 until 1920 he was en- 
gaged in similar activity at the Delta 
Laboratory, Arlington, N.J., where he 
worked on the acetylation of cellulose. In 
1920 he went to the Redpath Laboratory, 
Parlin, N.J., where investigations on the 
manufacture of photographic film were in 
progress. In 1925, he became director of 
research for the Du Pont-Pathe Manufac- 
turing Corp., at Parlin. When this or- 
ganization became Du Font's Photo 
Products Dept., he moved to Wilmington 
in the same capacity. He became tech- 
nical consultant to the department in 1942 
and technical adviser in 1946. He was 
named director of the development sec- 
tion in 1947. 

Dr. Sease's research on photographic 
emulsions included studies on precipita- 
tion conditions, the role of iodide and the 
nature of gelatins. He was particularly 
interested in grain size and in this con- 
nection worked on the formulation of de- 
velopers to control graininess. He wrote 
a number of technical articles on photo- 
graphic developments and was responsible 
for the development of a number of pa- 
tents in the photographic field. 

He was born near Leesville, S.C., in 
1888. He received his B.A. degree from 
Newberry College in 1908, was principal 
of a high school in South Carolina from 
1908 until going to Newberry College as an 
instructor in 1911. He was a fellow at 
Johns Hopkins from 1916 to 1917 and he 
received his Ph.D. from Johns Hopkins 
in 1917. He and Mrs. Sease now live at 
1010 Berkeley Rd., Wilmington 67, Del. 



New Members 



The following members have been added to the Society's rolls since those published last month. 
The designations of grades are the same as those used in the 1950 MEMBERSHIP DIRECTORY. 

Honorary (H) Fellow (F) Active (M) Associate (A) Student (S) 



Aragones, Daniel, Partner, Managing Di- 
rector, Laboratorio Cinefoto. Mail: 
Ave. Oral. Franco 426, Barcelona. (A) 

Bennett, Norman, American University. 
Mail: 10707 St. Margarets Way, Kensing- 
ton, Md. (S) 

Bischof, Wallace P., 3524 E. Anderson Ave., 
Albuquerque, N. M. (A) 

Bodnar, John, Head, International Cutting 



Dept., Twentieth Century-Fox Film Corp. 

Mail: 1210 Daniels Ave., Los Angeles 

35, Calif. (M) 
Bradish, Albert S., Vice-President in Charge 

of Production, Atlas Film Corp. Mail: 

1526 N. Harlem Ave., River Forest, 

111. (M) 
Bunnell, Ray F., Head, Electrical Dept., 

Warner Brothers Pictures, Inc. Mail: 



366 



900 E. Angeleno Ave., Burbank, Calif. 
(M) 

Clayton, Vincent E., Chief Engineer, Radio 
Service Corporation of Utah. Mail: 
1525 Browning Ave., Salt Lake City, 
Utah. (M) 

Coan, Alexander D., Television Advertising, 
Philbin, Brandon & Sargent, Inc. Mail: 
277 Park Ave., New York 17, N.Y. (A) 

Dance, Darrell A., Chief, Technical Services 
Branch, Motion Picture Division, U.S. 
Dept. of State. Mail: 15 Arcadia Rd., 
Apt. 21A, Hackensack, N.J. (M) 

Davis, Wesley J., Boston University. Mail: 
125 Homestead St., Boston 21, Mass. 
(S) 

de Gorter, Benjamin, Research Librarian, 
Technicolor Motion Picture Corp. Re- 
search Dept., 6311 Romaine St., Holly- 
wood 38, Calif. (A) 

Paris, Herbert C., General Manager, Tele 
Sales, Inc. 118 St. Clair Ave., N.E., Cleve- 
land, Ohio. (A) 

Fitzstephens, John J., Associate Instructor, 
New Inst. for Film and Television. Mail: 
315 W. 14 St., New York 14, N.Y. (A) 

Glubin, Samuel B., New Inst. for Film and 
Television. Mail: 231 Snediker Ave., 
Brooklyn 7, N.Y. (S) 

Gomez, Miguel A., New Inst. of Film and 
Television. Mail: 23 A Ulmer Dr., Brook- 
lyn, N.Y. (S) 

Gordon, Robert N., Manufacturing Engi- 
neer. Mail: 6009 W. Pico Blvd., Los 
Angeles 35, Calif. (A) 

Greenberg, Raymond, New Inst. for Film 
and Television. Mail: 149 Avenue C., 
New York 9, N.Y. (S) 

Jantzen, Charles A., Photographic Analysis 
Co. Mail: 582 E. Seventh St., Brooklyn, 
N.Y. (M) 

Johnston, Frank B., Sr., Chief Photographer, 
Philadelphia Inquirer. Mail: 1417 Dorset 
Lane, Philadelphia 31, Pa. (A) 

MacDonald, Joseph W., New Inst. for Film 
and Television. Mail: 2414 Sullivant 
Ave., Columbus 4, Ohio. (S) 

Mahoney, William J., Chief Audio Engineer 
and Owner, Cinecorders. Mail: 1730 
Kleemeier St., Cincinnati 5, Ohio. (A) 

Maloney, T. J., Producer-Director, KEYL- 
TV, Transit Tower, San Antonio, Tex. 
(M) 

Mclaughlin, Charles D., Projectionist, 
Southland Drive-In Theatres. Mail: 
5655H Huntington Dr., Los Angeles 32, 
Calif. (A) 

Miceli, Ernest, Film Editor and Experi- 
mental Cinematographer, WOR-TV. 
Mail: 1745 W. Ninth St., Brooklyn 23, 
N.Y. (A) 

Miller, Eugene S., Development Engineer, 
Eastman Kodak Co. Mail: 123 Heber- 
ton Rd., Rochester 9, N.Y. (M) 



Niehus, Murray H., Engineer, Cliffords 
Theatre Circuit. Mail: 313A King Wil- 
liam St., Adelaide, South Australia (A) 

Norbury, Alfred S., Engineering Aide, Corps 
of Army Engineers. Mail: 3526 Harri- 
son St., Kansas City 3, Mo. (M) 

Nordbye, R. B., Motion Picture Photog- 
rapher and Director, Great Lakes Prod., 
Inc. Mail: 208 North Hale, Wheaton, 
111. (A) 

Peltz, Leo G., Hollywood Sound Inst., Mail: 
1023 N. Edgemont, Los Angeles 27, Calif. 
(S) 

Searle, Milton H., Quality Control Engineer, 
Eastman Kodak Co., Inc. Mail: 216- 
10 47 Ave., Bayside, N.Y. (A) 

Stagnaro, John A., Chief Station Engineer, 
KECA-TV, American Broadcasting Co. 
Mail: 1105 N. Louise St., Glendale 7, 
Calif. (M) 

Stoddard, William C., Boston University 
School of Public Relations. Mail: Cochi- 
tuate Road, Wayland, Mass. (S) 

Tucker, Morris H., Technician, Columbia 
Broadcasting System. Mail: 1530 Archer 
Rd., Bronx 62, N.Y. (A) 

Wirth, Charles H., Project Engineer, Ranger- 
tone, Inc. Mail: 78 N. Spring Garden 
Ave., Nutley, N.J. (A) 

Zampari, Carlo, Studio Manager and Asso- 
ciate, Studios Vera Cruz, "Sao Bernardo 
do Campo," Sao Paulo, Brazil. (A) 

CHANGES IN GRADE 

Allen, Eugene S., Jr., Cameraman and 
Editor, Video Films. Mail: 870 New- 
port Ave., Detroit 15, Mich. (S) to (A) 

Brauer, Howard H., Chief Electronics Engi- 
neer, Bell & Howell Co. Mail: 1020 
Lawrence Ave., Chicago 40, 111. (A) to 
(M) 

Hufford, Robert G., Physicist, Eastman 
Kodak Co., 6706 Santa Monica Blvd., 
Hollywood 38, Calif. (A) to (M) 

Jones, Lt. Harold, Jr., Photography Director, 
U.S. Signal Corps. Mail: Apt. 9A, 2026 
Magill Dr., Fort Monmouth, N.J. (A) 
to(M) 

Mann, R. G., Chief Engineer, Pathe News, 
625 Madison Ave., New York 22, N.Y. 
(A) to (M) 

Rockar, Lt. William F., Signal Corps, 301st 
Signal Photo Co., Camp Gordon, Ga. 
(A) to (M) 

Svancara, V. J., Chief Sound Engineer 
(Motion Pictures), Wright-Patterson Air 
Force Base. Mail: 1228 Epworth Ave., 
Dayton 10, Ohio. (A) to (M) 

Tamer, James S., Photographer, Sandia 
Corp. Mail: 3216 A. St., Sandia Base, 
Albuquerque, N.M. (A) to (M) 

Zimmermann, August H., Engineer, DeLuxe 
Laboratories, Inc. Mail: 1090 Trafalgar 
St., West Englewood, N.J. (A) to (M) 



367 



New Products 



Further information about these items can be obtained directly from the addresses 
given. As in the case of technical papers, the Society is not responsible for manufac- 
turers' statements, and publication of news items does not constitute an endorsement. 

5 g horizontal, but a special combination 
is available for greater acceleration loads. 
A film-driven switch actuates an electric 
footage counter and also gives remote indi- 
cation of camera operation. Several types 
of motors are available for field or mobile 
use. Power requirements range from 10 w, 
d-c, to 115 w, a-c, depending on applica- 
tion. Frame registration is accurate to 
0.003 in., adequate for motion analysis and 
motion picture projection. An intermit- 
tent movement pin is used to transport 
and register film. 

The camera is 6 X 5 X 2^ in., with 
motors extending from 2 to 3 hi. on one 
side. The weight, with aircraft motor, 
loaded 400-ft magazine and lens, is 12 Ib. 
It is constructed to withstand tempera- 
tures ranging from -40 to 160F. 




A 35-mm recording camera to fill the 
needs of industrial and research analysts 
has been designed by A. P. Neyhart, 
Chairman of the SMPTE's Subcommittee 
on Industrial and Research Photography, 
and H. G. Cunningham, camera designer, 
Hollywood. The Automax, as the new 
instrument has been named, is being tooled 
for production in the form shown above by 
the Guild Laboratories, 6264 Sunset Blvd., 
Hollywood 28, Calif., with which Neyhart 
is associated. 

Neyhart is the designer of the camera for 
surgical photography which has been de- 
scribed in the JOURNAL (p. 747, June 1950). 

The Automax can be operated, locally or 
by remote control, at interval rates from 
one exposure per hour to five exposures per 
sec and at pre-set cine rates of from 10 to 
48 exposures per sec. Film exposure is 
the same for both interval and cine opera- 
tion. An external intervalometer is re- 
quired for automatic sequence operation. 
A 400-ft Mitchell magazine is standard in 
the present design but other capacities 
may be used. The standard design is for 
an acceleration range of 10 g vertical and 



With a new double-function, mercury 
arc-lamp power supply, manufactured by 
the Huggins Laboratories, 778 Hamilton 
Ave.., Menlo Park, Calif., either direct- 
current or single-flash operation of AH-6 
or BH-6 mercury-vapor arc lamps may be 
obtained from 115-v, 60-cps power. 
Direct-current operation provides steady 
light, while pulsed operation gives a 
brilliance about 200 tunes greater, with a 
duration of approximately 10 /usec. Either 
mode of operation can be selected from a 
single switch. With d-c setting, the power 
supply delivers 1 kw, 1.2 amp at 800 v. 
Open-circuit voltage of 1700 v is supplied 
for starting the lamp. Standard d-c ripple 
is about 5%, but lower values can be sup- 
plied in special units. With flash opera- 
tion, the 10 /isec pulse at a power of 
approximately 2.5 watt-sec is provided by 
a power capacitor discharging through the 
lamp by means of a thyratron-controlled 
spark gap. Maximum repetition rate is 
6 pulses per mm. 

The unit is mounted in a standard relay 
rack cabinet, and its over-all dimensions are 
22 X 31 X 15 in. The manufacturers 
suggest its application to high-speed, 
Schlieren, shadowgraph and interferom- 
eter photography. 



368 



Observer Reaction to Non-Simultaneous 
Presentation of Pictures 
and Associated Sound 

By Harold N. Christopher 



The results of experiments to determine observer reaction to non-simul- 
taneous presentation of pictures and associated sound are reported. Curves 
are presented for two groups of observers: (1) a conditioned or technical 
group, and (2) a group more nearly representative of motion picture or 
television audiences. The data given may be used to predict observer 
reaction to the lack of simultaneity between picture action and the re- 
sulting sound over a range of to 300 milliseconds. 



DUE TO THE DIFFERENCE in Velocity 
of light and sound in air, we, as ob- 
servers, perceive action and then hear 
the sounds resulting from the action. 
This has resulted in a natural condition- 
ing which causes the observer uncon- 
sciously to restore the action and sound 
to proper perspective. If, however, the 
delay in the sound path exceeds certain 
limits the observer not only becomes 
conscious of the delay but is likely to 
voice vigorous objections. Organiza- 
tions dealing with the recording, repro- 
duction or transmission of pictures and 
associated sound have recognized this 
and have found it necessary to maintain 
a reasonable degree of simultaneity be- 
tween picture actions and accompanying 
sounds. It is the intent of this paper 
to give the results of tests to determine 



Presented on October 20, 1950, at the 
Society's Convention at Lake Placid, 
N.Y., by Harold N. Christopher, Bell 
Telephone Laboratories, Murray Hill, 
N.J. 



when the lack of simultaneity between 
the picture and the sound of a televised 
subject or sound motion picture becomes 
noticeable and objectionable, thus per- 
mitting a quantitative evaluation of the 
phenomenon. 

The method employed was to record 
observer reaction to a series of presenta- 
tions of actions and sounds in which the 
difference between the time of arrival of 
the visual and aural stimuli was care- 
fully controlled. In order to facilitate 
the taking of data an arbitrarily deter- 
mined comment scale was employed 
and is indicated below. 

No. Comment 

1. Not perceptible 

2. Just perceptible 

3. Definitely perceptible 

4. Definitely perceptible but not 

objectionable 

5. Somewhat objectionable 

6. Definitely objectionable 

7. Unusable 



April 1951 Journal of the SMPTE Vol.56 



369 



D AVERAGE DELAY FOR 
A GIVEN COMMENT 

A AVERAGE COMMENT 1 
FOR A GIVEN DELAY 




50 100 150 200 250 300 350 400 45O 

DELAY IN MILLISECONDS 

Fig. 1. Mechanical-percussion tests, sound delayed; 
20 observers (nontechnical). 



Distributions showing the percentage 
of all observations for an indicated com- 
ment number or less, for a given sound 
delay, (or advance) for talking and 
striking type actions and the resulting 
sounds are presented. 

Summary of Results and Conclusions 

The results of these tests indicate 
that the ability of an observer to detect 
the lack of simultaneity between action 
and the resulting sound is a function of 
the type of action and whether the sound 
is delayed or advanced with respect to 
the action. When the sound follows 
the action (normal order), delays in ex- 
cess of about 100 milliseconds are likely 
to cause adverse comment. When the 
normal order is reversed and the sound 
precedes the action, 35 milliseconds ad- 
vance is about the maximum unlikely to 
be objectionable. The above conclusions 
are based on observer reaction to the 
more easily correlated striking motions 
and percussion-type sounds. For scenes 



of people in conversation and moving 
about, the correlation between picture 
action and sound becomes more difficult 
for the observer and the above limits 
can probably be doubled. 

Description of Test and Data 

The first test devised and used in this 
study employed a complete television 
chain and a tape recorder with a mov- 
able reproducing pole-piece. This array 
of equipment permitted the televising 
of a subject talking or striking an object 
and the delaying of the reproduction of 
the sound by known amounts of time by 
means of the tape recorder. As the 
test proceeded, numerous questions 
arose concerning the adequacy of the 
quality of the pictures and the sound 
channel. Both were improved consider- 
ably but appeared to have little effect 
on the results. Direct-viewing tests 
were then resorted to, eliminating first 
the television equipment and finally the 
tape-recording equipment. This was 



370 



April 1951 Journal of the SMPTE Vol. 56 




350 



12 5 10 20 30 40 50 60 70 80 90 95 98 99 

PER CENT OF OBSERVERS VOTING AS INDICATED FOR A GIVEN SOUND DELAY 

Fig. 2. Mechanical-percussion tests, sound delayed; 
20 observers (nontechnical). 



made possible by a simple mechanical 
device consisting of two cam-actuated 
felt hammers, one of which struck a 
dead microphone a sharp blow and re- 
turned rapidly to its striking position in 
full view of the observer. The second 
hammer, not visible to the observer, 
struck a live microphone causing a 
sharp popping sound in the sound re- 
producing system.* By indexing the 
second cam with respect to the first it 
was possible to produce at will a sound 
that was delayed or advanced with re- 

* The loudspeaker was located about 3 ft 
in front of the observer, slightly to his 
right but almost in line with the action, 
at about eye level. The striking ham- 
mer was approximately 8 ft in front of 
the observer. 



spect to the visible striking of the ham- 
mer. Only tests herein termed "me- 
chanical percussion" employed this de- 
vice and were made by determining ob- 
server reaction to a presentation of 
intermixed delays and advances. Tests 
other than mechanical percussion per- 
mitted an irregular presentation of de- 
lays only. The instruction to the ob- 
servers for all tests was to focus their 
attention on the picture action and vote 
their reaction to the lack of simultaneous 
presentation in terms of the numbered 
comments, a list of which was before 
them during the test. No time limit 
was imposed. A given condition was 
presented repeatedly, and changed only 
after the observer had commented. 
Two groups of observers were em- 



Harold N. Christopher: Reaction to Non- Simultaneity 



371 



COMMENT 

1. NOT PERCEPTIBLE 

2. JUST PERCEPTIBLE 

3. DEFINITELY PERCEPTIBLE 
DEFINITELY PERCEPTIBLE, 
BUT NOT OBJECTIONABLE 

5. SOMEWHAT OBJECTIONABLE 

6. DEFINITELY OBJECTIONABLE 

7. UNUSABLE 



4. 



50 



100 



2 

5 150 



200 



250 




12 5 10 20 30 40 50 60 70 80 90 95 98 99 

PER CENT OF OBSERVERS VOTING AS INDICATED FOR A GIVEN SOUND ADVANCE 

Fig. 3. Mechanical-percussion tests, sound advanced; 
10 observers (all engineers). 



ployed, the first consisted of a condi- 
tioned group, indicated as, "10 ob- 
servers, all engineers." In each test, 
however, the observers were not always 
the same 10 people. They were, in 
general, engineers familiar with prob- 
lems involving pictures and sound. 

The second group totaling 20 people, 
consisted of 4 engineers and 16 other 
observers, most of whom were drafts- 
men and included 5 women. This 
group although more nearly representa- 
tive of a picture audience was comprised 
of people probably above average in 
technical background and understand- 
ing. 

The observer comments for the me- 
chanical percussion tests are listed in 
Table I. An examination of these data 
indicates that both observer groups 
comment more severely when the sound 
occurs before the action and it also ap- 



pears that the 10-observer group is 
somewhat more critical (they comment 
more severely sooner) than the 20- 
observer group. The information or 
intelligence that appears desirable to 
extract from these data if possible is: 
(1) how severely would these groups 
react to a particular delay or advance in 
milliseconds; and (2) what delay would 
cause a given percentage of the observers 
to make the comment 4 or less. This 
would require two forms of processing, 
one in terms of average comment for a 
given delay and the second in terms of 
average delay for a given comment. 

Figure 1 plots the "20-observer, 
mechanical-percussion test, sound de- 
layed" data in the form of the matrix, 
in which the numbers at each data point 
correspond to the total vote for a given 
comment at that delay as indicated in 
Table I. The 50% points of the two 



372 



April 1951 Journal of the SMPTE Vol. 56 



' 



2 100 



150 



200 



250 



300 



350 



COMMENT 

1. NOT PERCEPTIBLE 
- 2. JUST PERCEPTIBLE 
3. DEFINITELY PERCEPTIBLE 
4. DEFINITELY PERCEPTIBLE, 
BUT NOT OBJECTIONABLE 
5. SOMEWHAT OBJECTIONABLE 
6. DEFINITELY OBJECTIONABLE 
7. UNUSABLE - 

3r 

c$X 


































x 


X 


X 


x 


x 


x 










C 


^ 


xC 


X 


X 


x 


x 


x 










X 
/ 


x 


?> 


j 


x 


x 


X 


x 

x 


x 










x 
x 


X 


X 


* 


rT 




X 


/ 


X 










X 


x 

x 


X 


x 


^ 


k 


X 














X 


x 

x 


x 


x 
X 


i 

x 


r 


















X 


x 
X 


X 





















12 5 10 20 30 40 50 60 70 80 90 95 98 99 

PER CENT OF OBSERVERS VOTING AS INDICATED FOR A GIVEN SOUND DELAY 

Fig. 4. Mechanical-percussion tests, sound delayed; 
10 observers (all engineers). 



methods of processing mentioned in the 
above paragraphs are indicated by the 
light dashed lines on the matrix. 
Through these lines the heavy straight 
line captioned 50% curve has been 
drawn. 

In a similar manner, although not 
shown in detail, the lines captioned 25% 
and 75% may be located. 

That the calculated 50% values de- 
viate in an irregular way from the 
straight-line 50% curve is evident and 
typical of all data presented here. It is 
felt, however, that the straight-line 50% 
curve as indicated is a practical repre- 
sentation of the observer's evaluation of 



the delay phenomena in terms of the 
comment scale, for either method of 
processing. It is also plain that straight 
line curves labeled 95% or 5% if 
shown on Figure 1 would fall in meager 
data areas and, therefore, would per- 
haps be less representative than the 
50%, 25% or 75% curves. 

Figure 2, a cross plot of the straight- 
line data shown on Fig. 1 results in a 
family of curves indicating how the 20- 
observer group reacted to delaying the 
sound with respect to the action for the 
mechanical-percussion test. 

Let us assume, for example, that we 
wish to predict how this group of people 



Harold N. Christopher: Reaction to Non- Simultaneity 



373 



Table I. Observer comments (see text for key to comment numbers). 

COMMENT NUMBERS FROM 20 OBSERVERS (iMON -TECHNICAL GROUP) 
SOUND DELAYED MECHANICAL PERCUSSION TEST 



DELAY 
MS 


1 
1 
1 

2 
1 
3 


1 
1 
2 
5 
3 


1 
1 
5 
5 
5 


1 
1 
1 

3 
3 


1 
1 
2 
5 
6 


1 
1 

2 

2 


1 
1 

2 
3 

2 
2 


b & 

1 
5 
5 


1111 
2121 
3223 


1 1 
2 2 

5 5 


1 1 
1 1 
1 4 


1 
1 
2 
4 
3 


1 3 
1 3 
2 3 
2 3 
2 4 



50 
100 
150 
200 
250 
300 
350 


6 


3543 


7 6 


5 6 


SOUND 


ADVANCED MECHANICAL PERCUSSION TEST 


ADVANCE 
MS 

50 
100 
150 
200 
250 
300 
350 


4 
6 


1 

2 
4 


1 

4 

7 


1 
1 
5 


3 
3 


(SAME 

1 1 
1 1 
1 1 


OBSERVERS AS ABOVE) 

111111 
1 11411 


1 1 

1 4 


1 1 
1 1 
J 7 


1 

2 


1 2 
2 2 
2 5 





























COMMENT NUMBERS FROM 10 OBSERVERS (ALL ENGINEERS) 



SOUND DELAYED MECHANICAL PERCUSSION TEST 



2111112211 



50 


3 


5 


3 


2 


1 


1 


1 


1 


3 


1 


100 


2 


5 


4 


5 


2 


2 


1 


5 


7 


5 


150 


3 


7 


7 


6 


2 


4 


4 


4 


7 


4 


200 


6 


7 


7 


5 


4 


5 


4 


7 


7 


6 


250 


7 


7 


7 


7 


3 


6 


6 


7 


7 


7 


300 


7 


7 


7 


7 


6 


7 


6 


7 


7 


7 


350 


7 


7 


7 


7 


5 


6 


6 


7 


7 


7 


SOUND ADVANCED MECHANICAL PERCUSSION 


TEST 



( SAME OBSERVERS AS ABOVE) 



1 1 t 



111111 



50 
100 
150 
200 
250 
300 
350 



1516154122 
6677576777 



77776 
77776 



76776 
76777 



77777 76 
7777776 



777 
777 



SOUND DELAYED TELEVISED TEST-H.G.FISHER TALKING 



DELAY 


j$ 


^ 




- 





' f. 


^ 0* 


MS 


^^ 


s y^ 


^ 


cT 


<T 


<r 


^ ^ * G 


40 


1 


2 


1 


1 


2 


2 


1 1 


100 


2 


1 


I 


1 


2 


1 


1 1 


125 


2 


2 


1 


1 


3 


1 


1 1 


150 


3 


1 


3 


2 


2 


1 


1 2 


200 


5 


2 


3 


2 


5 


3 


2 3 


250 


5 


3 5 


4 


3 


5 


4 ^ 


I 2 6 



300 



765656 



456 



SOUND DELAYED WINDOW TEST-H.G.FISHER TALKING 



40 
100 
125 
150 
200 
250 
300 



1 1 


1 


2 


2 


1 


4 




1 1 


2 


2 


1 


1 


4 




1 1 


4 


3 


2 


1 


4 




1 2 


3 


3 


3 


3 


5 




2 5 


3 


4 


1 


2 


4 




5 6 


5 


6 


5 


5 


6 ; 


> 2 


6 7 


4 


7 


5 


6 


5 ; 


I 2 



767 



would react to a delay of 100 milli- 
seconds in the sound track of a motion 
picture involving motions that result in 
percussion-type sounds. From Fig. 2 
we note that 87.5% would make the 
comment 4 or less (12.5% would vote 4 
or more), that 96% would vote comment 
5 or less, and 4% would comment 
more severely than comment 5. In 
other words, with this family of curves, 
if we know the delay we can determine 
or predict how many observers would 
object to the non-simultaneous presen- 
tation. 



Figures 3, 4 and 5 are to be interpreted 
as described above for Fig. 2 for the ob- 
server group and test conditions indi- 
cated in the figure captions. 

Data from two other tests are given in 
Table I. One test employed a com- 
plete television chain and tape recorder 
to produce the various delays, and hi the 
other the speaker was observed through 
a window in a sound-proofed panel, the 
sound being delayed by means of the 
tape recorder. 

These data, when processed as pre- 
viously described, resulted in distribu- 



374 



April 1951 Journal of the SMPTE Vol. 56 



COMMENT 

1. NOT PERCEPTIBLE 

2. JUST PERCEPTIBLE 

3. DEFINITELY PERCEPTIBLE 

4. DEFINITELY PERCEPTIBLE, 
BUT NOT OBJECTIONABLE 

5. SOMEWHAT OBJECTIONABLE 

6. DEFINITELY OBJECTIONABLE 

7. UNUSABLE 



100 



< 150 



200 




12 5 10 20 30 40 50 60 70 80 90 95 96 99 

PER CENT OF OBSERVERS VOTING AS INDICATED FOR A GIVEN SOUND ADVANCE 

Fig. 5. Mechanical-percussion tests, sound advanced; 
20 observers (nontechnical). 



50 



100 



150 



200 



250 



300 



COMMENT 

1. NOT PERCEPTIBLE 

2. JUST PERCEPTIBLE 

3. DEFINITELY PERCEPTIBLE 

4. DEFINITELY PERCEPTIBLE, 
BUT NOT OBJECTIONABLE 

5. SOMEWHAT OBJECTIONABLE 

6. DEFINITELY OBJECTIONABLE 

7. UNUSABLE 



350 




12 5 10 20 30 40 50 60 70 80 90 95 98 99 

PER CENT OF OBSERVERS VOTING AS INDICATED FOR A GIVEN SOUND DELAY 



Fig. 6. Average of talking tests; 10 observers (all engineers). Effect of delay 
in terms of comment scale. 



Harold N. Christopher: Reaction to Non- Simultaneity 



375 




40 



80 120 160 200 240 260 320 

DELAY OR ADVANCE IN MILLISECONDS 



360 



400 



Fig. 7. Non-simultaneous reproduction of pictures and sound; 

comparison of average data for 10-observer (technical) group 

and for 20-observer (nontechnical) group. 



COTTON MALLET, 
DIRECT VIEW 



- TELEVISED 
HAMMER TEST 



MECHANICAL 

PERCUSSION 

TEST ~~- 




COTTON MALLET, 
" WINDOW TEST 






X 



AVERAGE OF ALL 
PERCUSSION TESTS 



25 



50 



75 



100 125 150 175 200 225 250 275 300 
DELAY IN MILLISECONDS 



Fig. 8. Comparison of percussion data; 
10 observers (all engineers). 



376 



April 1951 Journal of the SMPTE Vol. 56 



tion curves that were for all practical 
purposes almost identical. The two 
sets of data were therefore combined 
and the resulting distributions are shown 
on Fig. 6. 

Discussion 

In order to compare the reaction of 
the two groups for the various test con- 
ditions a plot of the 50% values as 
found on Figs. 2, 3, 4, 5 and 6 is shown 
on Fig. 7. Here we have milliseconds 
delay or advance versus comments, and 
we can compare directly the average 
curves of the two groups for the various 
test conditions. From Fig. 7 it is seen 
that although the 10-observer group de- 
tects delays or advances before the 20- 
observer group, the curves for the two 
groups parallel each other. A constant 
difference of approximately one com- 
ment is indicated for the delay con- 
dition and 1.5 comments for the ad- 
vance condition. This points out a 
rather nominal difference between the 
average technical and the average non- 
technical observer. If this difference 
could be attributed to education, it is 
possible that repeat data on the 20-ob- 
server nontechnical group might more 
nearly approach the evaluation of the 
technical group. Another factor not dis- 
cussed but evident in the tabulations on 
Table I is that of audio-visual coordina- 
tion. The importance of this factor and 
the education factor cannot be deter- 
mined from the data presented here. 

Comparing the talking test with the 
mechanical-percussion test for the 10- 
observer group it will be noted that 
the average observer finds it more diffi- 
cult to detect the delay for the talking 
test and therefore appears more tolerant 
of delays that involve correlation of 
sounds with lip motions. The slope of 
the talking test curve indicates a some- 
what different evaluation of the delay 
phenomenon; the difference however 
appears too small to be significant. 

As mentioned briefly in the descrip- 



tion, tests other than direct-viewing 
mechanical-percussion and talking tests 
were made. Figure 8 shows, for com- 
parison purposes, four different percus- 
sion-type tests employing the 10-ob- 
server group. These curves show a re- 
markable similarity when one con- 
siders the following facts: (1) the 
greatly different presentation as indicated 
by the curve captions; and (2) that 
although 10 observers are indicated for 
each test, the observers were not always 
the same 10 people. 

Conclusions 

Observer reaction to non-simultane- 
ous reproduction of pictures and sounds 
has been rated subjectively by means of 
an arbitrarily determined scale of com- 
ments. 

Although a nominal difference is indi- 
cated in the evaluation of the phe- 
nomenon for technical and nontechnical 
observers, the subjective reactions of 
the two groups are almost exactly 
parallel. 

The average observer detects earlier 
and voices more vigorous objections to 
advances than delays. 

For delays, the objectionable effects of 
the more easily correlated actions and 
sounds (percussion type), though de- 
tected at considerably shorter delays 
than those between lip motions and the 
corresponding sounds, appear to grow at 
approximately the same rate. 

Differences in presentation that in- 
volved picture and sound quality have 
little or no bearing on the evaluation of 
the more easily detected delays. 

The repeated similarity of processed 
data when displayed or compared in the 
form of average curves (Fig. 8) sug- 
gests that the findings herein pre- 
sented, though determined from obser- 
vations of a relatively small number of 
people, would be changed or modified 
rather little had a greater number of ob- 
servers been employed. 



Harold N. Christopher: Reaction to Non-Simultaneity 



377 



The Television Cameraman 



By Rudy Bretz 



The requirements of the television medium and the unique design of tele- 
vision cameras have developed, in the best of television cameramen, oper- 
ators of unusual skill. A broad picture of the abilities and backgrounds of 
present-day television cameramen is presented, and a general compari- 
son is made between the television cameraman and the motion picture 
camera operator. The creative role of the cameraman and his relationship 
to the television director are explained in an outline of the stages of camera 
rehearsal. 



Handling the camera is a highly cre- 
ative job, and there is a tremen- 
dous difference between a good and a 
mediocre cameraman. The ability of a 
television cameraman depends on cer- 
tain basic abilities, but is also due in 
large measure to his attitude toward his 
job. This in turn seems to hinge pri- 
marily on the position of the camera- 
man within the station organization. 

About half the stations have been 
classifying cameramen as engineers. 
Not all of these require the cameraman 
to have a very thorough technical knowl- 
edge. At about half of these stations, 
the cameramen-engineers are assigned 
to camera operation and nothing more, 
and they are expected to be experts 
only in the art of camera handling. 



A contribution submitted January 10, 
1950, by Rudy Bretz, Television Con- 
sultant and Producer. This is part of a 
forthcoming book and is published by per- 
mission of McGraw-Hill Book Co., Inc. 
Critical discussion of this material is par- 
ticularly invited, either in the form of 
Letters to the Editor or by communica- 
tion directly to the author at Croton-on- 
Hudson, X.Y. 



The balance of these stations employ no 
cameramen as such, but apply a policy 
of rotation of the engineering personnel. 
An engineer may be assigned to the 
camera one week, to maintenance of 
equipment the next and as technical 
director the third. This is considered 
good management because it keeps the 
staff flexible; in case of illnesses or 
vacations it is possible to replace people 
easily, and more can be accomplished 
with a smaller staff. However, all 
engineers must then be well-trained 
technical men with a thorough knowl- 
edge of circuits and electronics. In 
most cases, such men do not particularly 
care for camera handling. Operating 
the camera calls for none of the special 
knowledge and skill which the technical 
man has acquired through his years of 
engineering study. At the same time, 
his background has been weak in the 
understanding of composition, picture 
showmanship and the visual arts. In 
many stations the engineers speak of the 
camera assignment as the "salt mine," 
endure it for as long as they must and 
make very little effort to contribute 
anything creative to the production at 



378 



Rudy Bretz: The Television Cameraman 



hand. In such a setup the producer has 
a much more difficult job doing a good 
show. 

The other half of television camera- 
men are classified in the production 
departments. They are responsible to 
the program director rather than to the 
chief engineer. Some stations even ro- 
tate personnel between cameras and 
other production jobs. A man may 
direct one show, take the camera for 
the following production, and then put in 
a stint as audio-console operator or pro- 
jectionist before the next show that he 
produces comes around. This is, of 
course, only possible in small towns 
where there are no iron-clad union juris- 
dictions. Such rotation means efficient 
station operation, and at the same time 
assures that the camera work will be 
as creative as possible. It also helps 
to eliminate social strata within the 
production organization. 

The job of cameraman is only one of 
a group known as operating jobs. Such 
duties as dolly-pushing, mike-boom 
operating, audio-console operating, the 
jobs of technical director, projectionist, 
record spinner, etc., are not strictly 
technical jobs. In none of these posi- 
tions does the operator have to under- 
stand more than the mechanics of oper- 
ation of his equipment. He is not 
called upon to repair or redesign, but 
only to operate, and skill of operation 
rather than engineering is required. A 
few television stations do classify all 
these jobs under the program depart- 
ment. An understanding of showman- 
ship is of greater value than a technical 
background in an operating job, and 
when production people are placed in 
these positions creative contribution is 
more likely to result. 

This is not to say, of course, that no 
engineers have any concept of show- 
manship. There are many engineers 
in television who, through particular 
backgrounds or extensive control-room 
experience in television or radio, have 
developed an understanding of the ele- 



ments of showmanship that would actu- 
ally qualify them as directors. The 
best of the "technical directors" fit this 
description. 

Aptitudes of the Successful Camera- 
man 

Whatever the cameraman's classifi- 
cation with the organization may be, he 
will become really good only if he has 
two essential aptitudes. The first is a 
sense of composition and the second is a 
well-developed manual coordination. 

The sense of composition comes only 
from long familiarity with a picture 
medium. A man who has been a still 
or motion picture photographer, or per- 
haps has worked on a picture magazine, 
has been thinking in terms of pictures 
and developing this sense. "Reading up" 
on composition doesn't help. It is not 
possible for the television cameraman, 
or director either, for that matter, to 
apply rules for composition while he is 
making pictures. He must be able 
to look at a picture, see what is wrong 
with its composition, and, by a condi- 
tioned reaction, as the psychologist 
would put it, unerringly make a quick 
adjustment to improve it. 

Manual dexterity and coordination 
come only to those who are endowed 
with the necessary aptitude. Just as 
it is impossible to teach some people to 
fly an airplane, so it is impossible to 
teach some people the smoothness and 
dexterity necessary to operate the tele- 
vision camera. A man who lacks the 
feeling for composition, but has coordi- 
nation, may learn the former in time; 
if he lacks the aptitude for physical 
coordination, he will never be a good 
television cameraman, no matter how 
finely developed his pictorial sense may 
be. 

Television and Motion Picture 
Cameramen 

Perhaps the unique nature of the 
television cameraman can best be ex- 
plained if he is compared with the near- 



Rudy Bretz : The Television Cameraman 



379 



est thing which existed before the ad- 
vent of the television medium the 
motion picture cameraman. The com- 
parison is not an easy one to make and 
possibly an accurate parallel can be 
drawn to only one phase of the film 
cameraman's work. For one thing, the 
television cameraman has no responsi- 
bility for lighting or exposure, or for 
picture quality except in regard to 
composition and smoothness of camera 
movement. He compares most closely 
to the operating cameraman in Holly- 
wood who, as an assistant to the direc- 
tor of photography, is concerned only 
with handling the camera. 

At first comparison the television 
cameraman is seen to have a lot more 
to do than his motion picture counter- 
part. The job of handling a motion 
picture camera is only a job of framing 
the picture. The camera operator 
controls the camera with the panning 
handle, pans the camera left and right, 
or tilts it up and down to keep a good 
composition. If the camera dollies in 
or out, or the subject moves toward or 
away from the camera, the focus must 
be adjusted on the lens barrel, but the 
cameraman has an assistant to help 
him in this operation. The assistant 
rides on the front of the camera dolly 
watching chalk marks on the floor. 
As the dolly wheel passes the 10-ft 
mark, he sets the lens to 10 ft; as it 
passes 8 ft, he has moved the focusing 
scale to 8, and so forth. If he cannot 
see the chalk marks on the floor, a 
second assistant walks alongside and 
whispers the distances in his ear. It 
is clear that this method of following 
focus can only be used when adequate 
rehearsal of each shot is possible. The 
television camera had to be designed 
so the cameraman himself, without pre- 
planning, can adjust focus to whatever 
motion is taking place within the scene. 

The television cameraman will often 
control his own camera movement as 
well. Many of the camera dollies used 
in television stations today may be oper- 



ated by one man and, in general, a good 
cameraman can usually work better 
alone than with an assistant. 

In the television camera, then, more 
aspects of the camera's operation are 
under one man's control. This gives 
him more work to do in one sense, but 
relieves him of a great burden at the 
same time the burden of coordinating 
the actions of two or more operators. 
When one man is operating, he is sub- 
ject to a certain possibility of human 
error. When two men are operating the 
same camera, however, the factor of 
human error is not merely doubled, it 
is multiplied by four. Each man's 
errors reflect upon the other. When as 
many as three operators must cooperate 
in the operating of a single camera 
(the counter-balanced-crane type of 
camera dolly requires three men), the 
factor of error may be increased by nine. 
Equipment such as this can be used to 
best advantage only when plenty of 
rehearsal time is available. 

The exception to this is the camera- 
man-dollyman team which has worked 
together so long that each man knows 
the other's reflexes and the two can oper- 
ate with a single accord even when cover- 
ing spontaneous action. 

The television cameraman develops 
more rapidly in his craft and reaches a 
high level of skill in much less time than 
it takes his motion picture counterpart. 
This is due to two factors: the actual 
amount of camera handling he does, 
and the fact that he can see his results 
as he works. 

In the usual motion picture studio, the 
largest part of shooting time is taken up 
in adjusting lights, rehearsing actors and 
in a thousand details of production. 
If an average day's shooting amounts 
to, say, five minutes of finished film, it 
can be assumed that the camera was 
in actual operation on takes and retakes, 
or on rehearsals before the takes, per- 
haps a maximum of twenty or thirty 
minutes. It was only during this time 
that the cameraman was practicing his 



380 



April 1951 Journal of the SMPTE Vol. 56 



skill. Like a musician who spends most 
of his day adjusting his piano, and only 
thirty minutes playing the instrument, 
he has acquired only a half hour of prac- 
tice toward the mastery of his instru- 
ment. " 

The television cameraman, on the 
other hand, works his eight-hour day in 
almost constant camera manipulation. 
The director invariably comes into the 
studio with more show to produce than 
he has rehearsal time for, and pushes 
the camera crew as hard and as fast 
as he possibly can. Of course, the 
cameraman is not working at top effi- 
ciency all this while if there are two or 
three cameras involved in the studio 
rehearsals, only one bears the entire 
burden of the production at any one 
time, while the others reposition for 
their next shots. There are moments, 
too, when the director is concerned with 
problems of acting, staging or audio, 
during which the cameraman can re- 
lax. It is safe to say, however, that a 
good three hours of his eight-hour day 
are devoted to actual camera handling. 
In comparison with the motion picture 
cameraman, then, he gets six times the 
experience in the same period of time. 

Not only does the television camera- 
man get more training, but he gets 
better training because he can see the 
results of his efforts as he works. The 
motion picture man is at a great dis- 
advantage in this respect. He must 
resort to complicated routines of meas- 
urement, using light meters and meas- 
uring tapes, simply because he cannot 
see whether the picture is well exposed 
or whether the subject is in focus. 
Likewise, the film cameraman who exe- 
cutes a dubious pan shot may think it 
is entirely acceptable until he sees it on 
the screen the next day. During that 
twenty-four hours, however, he has al- 
lowed a false impression to crystallize in 
his mind. He must find out his mistake 
and unlearn it before he can return and 
improve his technique. The television 
cameraman is under no such disadvan- 



tage. A poor shot is immediately evi- 
dent to him; he corrects the error at 
once, and the lesson has been learned. 

With these factors in mind, then, it is 
easy to understand why the best of tele- 
vision camera work is on such a high 
level. A good cameraman, after a year 
or two on the television camera, has 
learned his equipment so thoroughly 
that it is practically an extension of his 
own body. He can seemingly make the 
camera do anything and go anywhere, 
and do it smoothly and perfectly the 
first time. He has developed tech- 
niques of handling the cameras and the 
camera dollies which the manufacturers 
of the equipment never imagined. It 
is not correct to say that the best of 
television camera work is superior or 
even equal to the best of motion pic- 
ture shooting the film medium will 
probably always show superiority in 
production techniques. The flexibility 
of the television camera, however, and 
the ability of the cameraman to pro- 
duce complicated shots smoothly and 
without rehearsal, is something entirely 
new in camera handling, and in this 
the good television cameraman is far 
superior to his motion picture counter- 
part. 

Television camera techniques are be- 
ginning to influence motion picture 
production and will probably have a 
wide application in the studios where 
speed and efficiency in production are 
important. Methods of continuous 
shooting have been developed where 
several cameras operate at one time, 
repositioning between shots much in the 
manner of television cameras. One 
method utilizes a small industrial tele- 
vision camera which is coupled to each 
film camera enabling the director to 
watch the shots and direct the camera- 
man from a control position just as in 
television production. In motion pic- 
ture production of this type, the film 
cameraman is operating in the same 
manner as a television cameraman, al- 



Rudy Bretz: The Television Cameraman 



381 



though his equipment is somewhat dif- 
ferent. 

The recent development of the tele- 
vision recording technique, which makes 
it possible to film a television show off 
the face of a kinescope tube and distrib- 
ute it among television stations, just 
as any film is distributed, may well be- 
come a standard method of film produc- 
tion, at least of films for television use. 
Recent improvements in television re- 
cording technique indicate that the time 
is not distant when television cameras 
and studio equipment will be installed 
by film producers, and the motion pic- 
ture camera operator will have to learn 
the television cameraman's technique. 



Creativeness in Television Camera 
Work 

The television director is responsible 
for planning the creative use of the 
camera, although he may lean heavily 
on the advice of his cameramen or his 
technical director. 

Some stations, notably those oper- 
ated by NBC, use the "technical direc- 
tor" system, in which the technical di- 
rector operates as a kind of head 
cameraman. He will be familiar with 
the show rehearsals, and will have had 
a share in the planning of shots and 
camera angles. During studio rehears- 
al he is in charge of the operation and 
placement of cameras, and is usually the 
only one who gives directions to the 
cameramen. 

Proponents of this system see in it an 
analogy to the method of Hollywood 
production, in which each film has two 
directors, one of whom is the "Director 
of Photography" and is in charge of all 
technical aspects of the production, 
while the other who carries the title of 
"Director" concerns himself largely 
with the broader problems of staging 
and acting. One director, who has di- 
rected programs on many stations, has 
said that working under the NBC sys- 
tem is like having a twin directing the 



program with you. During rehearsal 
much time can be saved. When in the 
course of rehearsal a stopping point is 
reached, the director can go out on the 
studio floor, make corrections in the 
action and come back to the control 
room to find all the camera changes 
made and everything ready to go again. 
Of course, this is predicated on the ability 
of the technical director. He must have 
almost as good a background as the 
director himself. He must be primar- 
ily a showman, not an engineer. Where 
this method is in use, however, the job 
of technical director is always an engi- 
neering job. The technical director 
must be in charge of the camera and 
control-room crew, and for this reason 
must be a superior engineer. Men 
who can fulfill all these requirements 
are rare or cannot be found for the sala- 
ries that are offered. 

The technical-director system breaks 
down when an inexperienced man is on 
the job, or when an "ad-lib" show is 
attempted. The writer has observed a 
green director, a green technical director 
and green cameraman attempt to use 
this method; and the results were 
miserable. The director would give a 
camera order, and the technical director 
would garble it a little in relaying it to 
the cameraman, since he had no clear 
idea of what was meant. Then the 
cameraman would do the wrong thing. 
"No! I didn't mean that!" the director 
would tell the technical director. "I 
meant so-and-so!" This was again re- 
layed to the cameraman. This time the 
cameraman would make an error. When 
it was all finally straightened out and 
everyone knew what it was the director 
wanted, it would turn out to be some- 
thing that couldn't be done anyway be- 
cause of some technical factor which no 
one had anticipated. 

During an ad-lib show, camera and 
cutting instructions must be given very 
rapidly and acted upon immediately, or 
action is lost. Sudden instructions to 
the cameramen cannot originate in the 



382 



April 1951 Journal of the SMPTE Vol. 56 



director in these cases, since by the time 
they are relayed through the technical 
director, the moment has passed. To be 
sure, the technical director himself may 
make the sudden decisions; but he is 
acting then in the capacity of director, 
which very few technical directors are 
able to do, or would be allowed to do. A 
good technical director could assist the 
director by keeping one step ahead of 
him. He could so engineer the cameras 
that the director would have a variety 
of shots to choose from in calling takes, 
and one camera at least would always 
have a good shot, well framed, from the 
proper angle to show the action, and 
ready at the right time. However, there 
is some question as to whether this could 
not be done just as well or more easily 
by the director himself. It is a general 
opinion that for ad-lib shows the tech- 
nical-director system does not work. 
Since most television stations must do 
the ad-lib type of production (and prac- 
tically all remote pick-ups fall into this 
category) most stations (90%) have de- 
cided against this method. 

A good compromise is achieved in 
some stations, and some network studios 
where both the director and the tech- 
nical director may talk to the camera- 
men at any time. This makes quick de- 
cisions possible and at the same time 
provides a two-director team for the re- 
hearsal and production of the show. 
After operating under this joint system 
many people have observed that the 
usual method, whereby the technical 
director simply operates the switching 
system under the director's command, 
wastes the capabilities of this individual 
who could be assisting the director at 
the same time. 

From script to screen the production 
may go through many stages, or few, 
depending on the complexity of the pro- 
duction, and the time allowed for re- 
hearsal. Commercial dramatic pro- 
grams usually enjoy a rehearsal ratio of 
10 hr of rehearsal with cameras to 1 hr of 
.air time. Sometimes important shows 



have rehearsed with studio facilities at a 
ratio of 15 or 20 to 1. Under these opti- 
mum conditions, the following stages 
may be observed. 

Stage IThe Paper Stage. Detailed 
preplanning is absolutely vital to tele- 
vision production. In the paper stage, 
the director works with floor plans and 
shot plotters; he makes little sketches 
on the script of what the shots should 
look like ; he visualizes the positions of 
the cameras in the studio; draws them 
on the floor plans for every shot, and in- 
sures that everything he visualizes is 
practical and will work. 

Stage 2 Outside Rehearsal. Rehears- 
als of complicated shows always begin 
outside the studio in a rehearsal hall or 
some suitable space. The plan of the 
studio sets is marked off on the floor; 
chairs or other furniture are used to 
simulate sets and props, and the per- 
formers get a good idea of the space in 
which they are to work. Here the 
director will move about, taking the 
place of one camera and then the other, 
as he views each shot from the position 
of the camera that will take it. Some 
directors use a portable view-finder with 
lenses, which will give them the field of 
view of the television camera lenses. 
This instrument is available on the 
market at a rather high figure. Other 
directors use homemade view-finders, 
frame viewers or a simple shoe box finder 
with a cut-out mask. Most directors, 
however, frame a picture with their own 
hands. Some use their arms, some their 
fingers, but the result is the same an 
easier visualization of a picture within a 
3x4 frame. (See illustration on fol- 
lowing page.) 

If it is possible economically, the ideal 
thing is to show the cameramen an en- 
tire rehearsal of a show outside the 
studio before the first use of cameras on 
the studio floor. Studio rehearsal with 
facilities is, to a large extent, a period 
for briefing the cameramen, the floor 
manager, the stage crew and the con- 
trol-room personnel on the many aspects 



Rudy Bretz: The Television Cameraman 



383 




Several methods which directors use to help them plan camera shots. 



of the show which the director has pre- 
viously worked out on paper and with 
the cast. If the cameramen have seen 
the show before rehearsal, much of this 
time can be saved. Further, the camera- 
men are authorities on the problems of 
space and traffic on the studio floor and 
can spot difficulties of which the director 
may be unaware. And finally, the crea- 
tive mind of the cameraman, and his 
own powers of visualization, are a great 
help to the director in this planning 
stage. 

Stage 3 Dry-Run. Many directors 
prefer to use their first hour or so 
of studio rehearsal for a dry-run, that 
is, to walk through the show from be- 
ginning to end without using the elec- 
tronic facilities at all, working with the 
cameramen and crew on the studio 
floor. Positions of cameras and angles 
of view are more easily visualized here 
than in either the paper stage or in the 
outside rehearsal stage of development. 

Stage 4 Rough Run-Through. A 
fourth stage is sometimes added here a 
straight run-through of the entire show, 
paying no attention to all the mistakes, 
rough places and problems that turn up. 
This can be very valuable for the crew, 
especially for the stagehands who must 
make scene changes or work props dur- 
ing the show. Only a complete run- 
through with cameras can give them a 
total picture of the show, and without 



it they will be somewhat confused unti! 
the last dress rehearsal puts all the 
pieces together for them. 

Stage 5 Work-Through (Stop-Start). 
This stage is the stop and start or "work- 
through" which will take up the major 
portion of the rehearsal period. The 
director works through the show, stop- 
ping whenever necessary and working 
out all his problems as he comes to 
them. It is during this rehearsal that 
all the fine details of camera work will 
be set, and the cast and cameras will be 
coordinated. 

Stage 6 Run-Through. The sixth 
stage is the final run-through which is 
done preferably without stopping. Some 
directors will run through the show as 
many times as possible before air time, 
others may rest content with one good 
dress rehearsal and spend the remaining 
time working on difficult sequences. 
More often, there is time for neither, and 
sometimes a show is worked through so- 
close to air time that there is not suffi- 
cient time for a complete run-through 
at the last. Most of the small stations, 
when they attempt dramatic shows or 
other complicated types of programs, 
cannot allocate sufficient rehearsal time 
for all these stages. In such cases, all 
the steps are eliminated except the paper 
stage, the outside rehearsal and the 
work-through. 



384 



April 1951 Journal of the SMPTE Vol. 56 



Opinion is divided as to how much 
responsibility should be vested in the 
cameraman for finding the right shot at 
the proper time. In the case of the unre- 
hearsed show where there is no set se- 
quence of shots, the cameraman is 
usually relied upon to "hunt for shots" 
when he is off the air. The director may 
look at a shot the cameraman has found 
and say: "No, I don't want to use 
that/' or "that's good, give it to me 
again when I tell you," or he may switch 
it immediately into the program. 

At the opposite end of the scale is the 
method of operation in which the 
cameraman makes no move at all, ex- 
cept the very obvious, without instruc- 
tions from the control room. It is a gen- 
erally accepted principle that a camera- 
man should operate like this while his 
camera is on the air, but most stations 
give him greater freedom and more re- 
sponsibility between shots. 

In the case of the scripted and re- 
hearsed show, the cameraman will al- 
ways be supplied with cues from the 
control room to remind him of his next 
shot each time he is switched off the 



program line. In many studios, how- 
ever, he is expected to take the major 
responsibility, and will keep a cue sheet 
on the camera listing each shot as it 
becomes established in rehearsal. He 
will often mark the studio floor so he 
can find the exact camera position that 
was established in rehearsal for each 
shot. 

Some of the better cameramen are 
strongly opposed to this method, how- 
ever; they feel that the important thing 
is the shot, not the camera position, and 
since the actors' positions may vary be- 
tween rehearsal and air, the camera may 
sometimes be on the right mark and not 
have the proper shot at all. This same 
principle is relevant to the calling of 
lenses. Some directors like to specify 
the lens that will be used on each shot, 
and call for that lens during the show. 
These cameramen feel that much more 
flexibility should be possible, and that 
the cameraman need only be reminded 
of the shot he is to take and then al- 
lowed to find it, using his own discretion 
as to the necessary lens or camera posi- 
tion. 



Rudy Bretz: The Television Cameraman 



385 



A Simplified Index for Color 
of Illuminants 

By Frank F. Crandell, Karl Freund and Lars Moen 



A new improved unit for the trichromatic measurement and description 
of illuminants is presented. Described are: methods of derivation; rela- 
tion to Kelvin color temperature scale; application to color film, filters and 
light sources; and a three-color instrument for measurement of light 
sources using this new Spectra Index unit. 



ONE YEAR AGO, at the Society's Con- 
vention, a paper was presented on 
"The Effects of Color Temperature on 
Motion Picture Production." 1 It was 
pointed out at that time that for prac- 
tically as long as color photography has 
existed, photographers and cinematog- 
raphers have been plagued by the prob- 
lem of color balance between the sensi- 
tized material employed and the light 
source to which it was exposed in making 
a picture. 

Some illuminants, such as flash, are 
quite constant in color. Incandescent 
light is variable but can be held within 
fair tolerances. Some sources are un- 
suitable for existing color films, though 
reasonably constant in hue. 

Daylight, however, is quite another 
story. The amounts of red, green and 
blue in daylight change with the hour, 
the season, the altitude and latitude, 
with the state of the sky and the weather, 
with atmospheric contamination in a 



Presented on October 17, 1950, at the 
Society's Convention at Lake Placid, N.Y., 
(read by Norwood L. Simmons), by Frank 
F. Crandell, Karl Freund and the late 
Lars Moen, Photo Research Corp., 127 W. 
Alameda Ave., Burbank, Calif. 



word, with so many variables that no 
table and no computer could possibly 
cope with them. As for the human eye, 
it is a notoriously poor judge of illumi- 
nants because of its elastic power of adap- 
tation. 

In a confused situation such as this, 
no one will dispute the need of two 
specific things : an adequate instrument 
for the measurement of the spectral 
energy distribution, or color balance, of 
any given illuminant, and a convenient 
unit or index number in which the read- 
ings of that instrument can be expressed. 

The solution of this problem which 
has existed until the present is well 
known to most color photographers. 
For lack of anything better, a unit was 
borrowed from the illuminating engi- 
neers Kelvin color temperature. This 
is the temperature on the Kelvin or ab- 
solute temperature scale to which a black 
body would have to be heated to give off 
light of the color in question. 

The unfortunate shortcoming of 
"color temperature" is that, while it 
may be quite accurate in dealing with 
incandescent light sources (which are 
virtually "black bodies"), it is wholly 
inadequate in dealing with daylight and 
most other artificial illuminants. The 



386 



April 1951 Journal of the SMPTE Vol. 56 



reason is simple. The Kelvin scale is 
adequate for the ratio of any given two 
colors in the spectrum (say, for example, 
the middle of the red and the middle of 
the blue) but for any given third color 
there is only one possible value at a 
given ratio of the other two. For ex- 
ample, for any given blue-red ratio, 
green can have only one value, as ex- 
pressed in Kelvin color temperature. 
Since any illuminant other than incan- 
descent lamps is highly likely to have 
more green or less green than a black 
body, color temperature fails to describe 
it. 

Since incandescent lamps commonly 
used for color photography can be rated 
on the Kelvin scale, the practice has 
grown up, rather haphazardly, of mark- 
ing color film as Type A (3400 K), 
Type B (3200 ' K), and "Daylight" 
(Kelvin temperature not specified, 
though usually assumed to be 5900 or 
6100 K). 

This has worked reasonably well in 
practice, under perfectly normal condi- 
tions. However, many kinds of "day- 
light" prove to be something quite 
other than what the film was balanced 
for by the manufacturer. 

When it comes to describing the 
properties of filters, Kelvin temperature 
breaks down completely. If a given 
correction filter, for example, alters 
color temperature 100 with incandes- 
cent lamps, the same filter at daylight 
temperatures may have an effect 
amounting to three or four hundred 
degrees! 

The first step toward clarification of 
this muddled situation was the develop- 
ment of a photoelectric instrument 
which would enable the photographer to 
measure color temperature instead of 
guessing it the Spectra Color Tempera- 
ture Meter. However, this still left the 
uncertainty as to the green content of a 
light source, and made it necessary to 
consult a table for the selection of a 
color temperature altering filter. 

For more than a year, therefore, ex- 



tensive work has been done on the de- 
velopment of a simple system to de- 
scribe and interrelate the properties of 
light sources, color film and color cor- 
rection filters by the use of an index 
number, making description of color of 
the illuminant as simple as setting an 
exposure meter with an ASA film speed 
index. 

The first step was the development of 
a new Spectra, no larger than the pre- 
vious instrument, which would measure 
both the blue-red ratio and also the 
green-red ratio, of any given light source. 
This is called the Spectra Three-Color 
Meter (see Fig. 1). The next step was 
to find a system for the calibration of 
these two scales that would be simpler, 
more rational and more informative 
than the Kelvin scale. 

A tentative proposal for such a sys- 
tem was put forward in the paper pre- 
viously cited, 1 in order to sound out 
industry opinion on the subject. Com- 
ments have been received from scien- 
tists, from manufacturers, from camera- 
men and photographers, and from illu- 
mination engineers. Many of their sug- 
gestions have been extremely helpful 
and have been adopted; all were heart- 
ily in favor of such a rational system; 
none favored retention of the clumsy 
Kelvin temperature scale. 

The results of all this have been 
incorporated in a new index which de- 
scribes the properties of a light source, a 
color film or a correction filter, known as 
the SI or Spectra Index.* This index is 
derived directly from the mathematical 

* This system of indices for light, film and 
filters is referred to as the Spectra Index 
(SI) system and where no ambiguity 
results, Spectra Index (SI) may be used 
for any one of the individual values. 
However, where the distinction between 
light, film and filter is to be denoted, the 
light index is referred to as Spectra Dis- 
tribution Index (SDI), the film index as 
the Spectra Sensitivity Index (SSI) and 
the filter index as Spectra Transmission 
Index (STI). 



Crandell, Freund and Moen: Illuminant Color Index 



387 




Fig. 1. The new Three Color 
Spectra Meter. 

It reads both the blue-red and green-red 
balance of the prevailing illumination so 
that correction can be made for illuminants 
that fall off the black-body locus, as well 
as those that are on the locus. 

properties of black-body radiation, f and 
can be duplicated by any manufacturer 
in any part of the world. It is felt that 
this step is as important in clarifying a 
muddled situation as was the original 
introduction of the Weston film-speed 



t In the derivation of the units in the 
Spectra Index system, use is made of the 
complete spectral distribution of a black 
body at a given Kelvin temperature and 
not just to its visual appearance or "color 
temperature." 2 



number for use with photoelectric ex- 
posure meters. 

In the paper presented a year ago, it 
was proposed that the index be derived 
from the logarithms of the readings ob- 
tained through a special set of standard 
filters a method which resulted in a 
straight-line locus for black-body radia- 
tion at all useful Kelvin temperatures, 
with equal divisions for equal differ- 
ences of temperature, provided that the 
latter was expressed in Micro-Recipro- 
cal Degrees (MRD). 

However, it has been pointed out 
that the artifice of the standard filters 
can be dispensed with, and the index 
derived directly from the mathematical 
properties of black-body radiation. If 
we take Wien's law for black-body radia- 
tion: 



J\ = 



e c z /xr 



where Ci and <7 2 are the radiation con- 
stants; 

Jx is the energy at a given wavelength, 
X; and 

T is the temperature on the Kelvin or 
absolute scale ; 

and take the log of the ratio of the 
energy at two wavelengths Xi and X2 
then 



where A = 5 log e (^ 
\Ai> 

and B = C 2 ( r-\ The result- 

\A 2 Ai/ 

ing graph of log e l against (the re- 
ciprocal of the Kelvin temperature) is 
a straight line. 

Carrying this line of reasoning a step 
farther, it is evident that if the log 
ratios of one pair of spectral lines, 
plotted against reciprocal Kelvin tem- 
peratures, yield a straight-line graph, 
then the log ratios of two pairs of lines, 
plotted against each other, will also 



388 



April 1951 Journal of the SMPTE Vol. 56 



yield a straight line, the slope of which 
may be considered the vector of the rate 
of change of the two sets of ratios. 

This is what has been done in setting 
up the scale of Spectra Index values. 
Since Wien's law is substantially valid 
in the region from 2000 to 10,000 K, it 
may be considered suitable for dealing 
with all light sources likely ever to be 
used in color photography. If desired, 
a very slight departure from linearity 
in the upper part of the scales will bring 
the reading into line with exact tem- 
peratures according to Planck's law. 

However, taking advantage of the 
useful mathematical properties of the 
straight-line locus, two important 
changes have been made in the index 
since it was first proposed. First, the 
logs of the blue-red and green-red ratios 
have been multiplied by arbitrary con- 
stants, so chosen that a difference of 10 
MUD in a light source will result in a 
difference of exactly one unit on the 
blue-red scale and of one-half unit on 
the green-red scale. (The reason for the 
use of the one-half unit will appear in 
the discussion of correction filters, later 
in this paper.) Second, for further con- 
venience, a constant has been added to 
both numbers to bring the index of a 
3200 K light source to exactly zero on 
both scales. 

The value of these changes is obvious. 
Since 10 MRD is generally accepted as 
the just perceptible difference which will 
create a visible difference in the result- 
ing color reproduction, the tolerance at 
any point in the scale becomes one-half 
unit. Since all light sources used in 
color photography have a color tempera- 
ture of 3200 K, or higher, only positive 
values will normally be encountered. 
This eliminates the possible confusion 
of negative indices for illuminants be- 
low 4000 K, and positive indices for 
values above, as first proposed. 

The formulas for the conversion of 
color temperatures to the Spectra Dis- 
tribution Index and vice versa are ex- 
tremely simple. They are : 



SDIu-R = 31.25 - 
SDI G -R = 15.63 - 



100,000 
T 

50,000 



T = 



100,000 



31.25 - SDI B -R 

50,000 
15.63 - SDI G -R 

in which 

SDIu-R = Blue-Red Spectra Index 
SDIa-R = Green-Red Spectra Index 
T = Kelvin temperature 

In practice, the values for the two 
ratios are combined, with the blue-red 
figure coming first; if the B-R value is 
5.5 and the G-R value is 2.8, the com- 
plete SDI is 5.5/2.8. It will be noted 
that for equivalent black-body radiation 
the Gr-R index must always be one-half 
of the B-R index (rounded off to the 
nearest tenth.) If this is not the case, 
we immediately know that the illumi- 
nant in question is off the black-body 
locus. 

The application of the foregoing to 
color film and to illuminants is immedi- 
ately apparent. The SDI of an illumi- 
nant tells us all we need to know of its 
properties for trichromatic photog- 
raphy; the SSI of a color film is simply 
a statement of the illuminant color 
which will yield the best balanced re- 
production on that particular film or 
coating. This still leaves the question 
of correction filters to bring the two 
into balance. This problem has been 
met in a way which is believed to de- 
serve somewhat extended treatment, 
since it represents a more comprehensive 
and systematic approach to the question 
than any known previous effort. 

Consider first the graph shown in Fig. 
2 which is the graph of light sources 
shown as Fig. 10 in the July, 1950, 
JOURNAL paper 1 now redrafted to fit 
the new Spectra Index. The illumi- 
nants, from A to M, are listed in Table I. 
This takes in the color space bounded 



Crandell, Freund and Moen : Illuminant Color Index 



389 



Table I 





Illuminant 


Simplified 
Filter SI or 
Color Designation* STIf 


A. 


Studio broadside 


4 steps too yellow 










2 steps too magenta 


4T 


4/2 


B. 


170 M-R lamp with 


3 steps too yellow 








5070 (Y-l) glass 


1 step too magenta 


3T 


3/1.5 


C. 


Noon sunlight 


3 steps too yellow 










1 step too magenta 


3T 


3/1.5 


D. 


Whitelite 6300 


1 step too yellow 


1 T 


1/0.5 






1 . 5 step too magenta 


1 G 


1/0 


E. 


Daylight on horizontal 


Correct 









plane; fairly clear 








F. 


Same; clear 


Correct 








G. 


Sun outside the atmos- 


1 step too blue 


1 S 


-1/-0.5 




phere 








H. 


Graf AC H-I Arc 


1 step too blue 


IS 


-1/-0.5 


I. 


Complete overcast 


1 step too blue 


1 S 


-1/-0.5 


J. 


Whitelite 7100 


1 step too blue 


1 S 


-1/-0.5 






1 step too magenta 


1 G 


0/1 


K. 


Illuminant C 


1 step too blue 


1 S 


-1/-0.5 


L. 


Sunshine Arc, white 


9 steps too blue 


6S 


-6/-3 




flame 


2 steps too magenta 


3S 


-3/-1.5 








2G 


0/2 


M. 


North sky, clear 


9 steps too blue 


6S 


-6/-3 






6 steps too green 


3S 


-3/-1.5 








1 M 


0/-1 



See footnote on p. 395. 



t See footnote on p. 388. 



by Spectra indices 8 to 24 in the blue- 
red and 1 to 13 in the green-red. For 
purposes of orientation, this embraces 
roughly the color temperatures from 
4000 to 15,000 (somewhat more in the 
green-red), and thus takes in any pos- 
sible light source which might be cor- 
rected for use with daylight type color 
film. 

This graph may be considered rather 
similar in its properties to the UCS 
color triangle. As we move upward on 
this chart, light becomes more greenish ; 
as we move to the right, it becomes more 
bluish.* 



* On Judd's Uniform-Chromaticity-Scale 
(UCS) triangle our two axes would 
correspond more closely to a line from 
the right (blue) corner to the middle of 
the left side (yellow or minus blue) as 
the B/R axis and the line from the top 



Now, let us set ourselves the hypo- 
thetical problem of correcting any illu- 
minant which might be encountered 
within this color space so that it will give 
a balanced result on daylight-t3 r pe color 
film. If the color of illuminants were a 
random matter, and we were likely to 

(green) to the middle of the bottom 
(minus green or magenta) for the G/R 
axis. There have been many proposals 
recommending two axes more or less in 
the blue-yellow, green-minus-green di- 
rections: Adams 3 uniform chromatic- 
ity system, Hunter's 4 ' 5 alpha-beta chro- 
maticity system, and Robertson and 
Milligan's 6 yellowness-greenness sys- 
tem, to name only three. They seem to 
have in common at least approximately 
uniform chromaticity scales with neutral 
gray as their center and an ease in vis- 
ualizing the hue represented by given 
coordinates. 



390 



April 1951 Journal of the SMPTE Vol.56 



19 
12 
II 
10 

9 
O 

LjJ A 

7 

z 

LJ 6 
LJ 
CC 
O 5 

4 
3 
2 

I 



o * 
> ' o 
o J 



A O 
O O 
O 

f 



10 II 12 13 14 15 16 17 18 19 20 21 

BLUE - RED 



22 29 24 



Fig. 2. Thirteen widely assorted illuminants plotted 
in Spectra Index coordinates. 

The diagonal line is the black-body locus. This graph has been redrawn, in terms of the 

revised Spectra Index values, from Fig. 10 in the Authors' paper in the July, 1950, 

JOURNAL, p. 85. The illuminants, A to M, are identified in Table I, p. 390. 



encounter in practice light sources which 
fell anywhere within the diagram, then 
the simplest solution of the problem 
would be a set of niters which shifted 
the balance along rectangular coordi- 
nates, i.e., a plus- or minus-blue series 
and a plus- or minus-green series. 

However, the color of illuminants is 
not a random matter. Thirteen real 
light sources, both natural and artificial, 
have been plotted in Fig. 2, marked A 
to M. It will be noted that these light 
sources tend very strongly to group 
along the black-body locus. Expressed 
in another way, the similarities of real 
illuminants to black-body radiation are 
greater than the differences. It will be 
noted that two quadrants of the chart, 
the upper left and the lower right, which 



are farthest from the black-body locus, 
have no illuminants plotted in them at 
all. For obvious psychological and 
technical reasons, this is likely to con- 
tinue to be true. The only illuminants 
which would fall far from the black-body 
locus would be highly colored sources, 
such as the sodium lamp or the old 
mercury vapor tube, which are not 
suitable for color photography with any 
possible degree of correction. 

This leads to an important conclusion 
regarding the problem of correction 
niters. Since all illuminants tend to be 
in the neighborhood of the black-body 
locus, there are obvious advantages in 
having one set of correction niters so 
designed that it shifts the hue of an 
illuminant in the direction parallel to the 



Crandell, Freund and Moen: Illuminant Color Index 



391 



12 13 14 15 16 17 18 19 20 21 22 29 24 




Fig. 3A. The same illuminants as those shown in Fig. 2, with the addition 
of a grid representing the coordinates of the four series of 
correction filters, salmon, turquoise, green and magenta. 

This grid is not static, but is always shifted so that its O-O point is located at the Spectra 
Index of the light source being used, and the number of squares that must be traversed 
to arrive at the SDI of the film (for example, 6000 K, or SDI 14.6/7.3, indicates the 
filter required). In practice, the entire operation would be carried out, not by means 
of a graph, but by setting the computer which is illustrated elsewhere. 



black-body locus. In a large number of 
cases, the total required correction can 
then be made with a single filter. 

For this purpose, we require filters 
with sloping rather than sharp absorp- 
tions. Instead of pure yellow (minus- 
blue) we must use a pinkish yellow 
which, for each unit of blue which it 
absorbs will absorb half a unit of green. 
Instead of a blue, we must use a more 
greenish hue, having a half-unit of green 
with each unit of blue. Such a series, of 
course, is substantially what has been 
used in the past for the correction of 
color temperature. 



This takes care of the blue-red cor- 
rection. There remains the question of 
any additional green-red correction 
which may be required. Superficially, 
it might well be assumed that the logical 
method of accomplishing this would be 
to use a series of greenish and pinkish 
filters which would shift illuminant 
color in a direction perpendicular to the 
black-body locus. Thus, the filters 
would be evenly spaced along a series of 
rectangular coordinates with the entire 
grid at an oblique angle. 

However, a little reflection will show 
that the logic of this is only apparent, 



392 



April 1951 Journal of the SMPTE Vol. 56 




Fig. 3B. Spectral transmission curves 
for CT-1T through CT-6T filters. 

These filters increase the blue-red ratio 
twice as much as they increase the green- 
red ratio, so they correct in the direction of 
increasing black-body temperature. 



Fig. 3C. Spectral transmission curves 
for CT-1S through CT-6S filters. 

These filters decrease the blue-red ratio 
twice as much as they decrease the green- 
red ratio, so they correct in the direction of 
decreasing black-body temperature. 



Fig. 3D. Spectral transmission curve 
for GC-3G filter. 

This filter reduces blue and red equally, so 
increases the green-red ratio only. 



Fig. 3E. Spectral transmission curve 
for GC-3M filter. 

This filter reduces principally green leav- 
ing blue-red ratio unchanged but reduces 
the green-red ratio. Note that GC-3G 
and GC-3M combine to form a photo- 
graphic neutral. 



and that the drawbacks would outweigh 
the advantages. Our yellowish and 
bluish filters affect both the blue-red 
and the green-red ratios, which is correct 
and desirable, since it keeps the shift 
parallel with the black-body locus. 
However, if we were to use pinkish and 



greenish filters which shifted the hue at 
right angles to the locus, then these 
filters would also affect both ratios. 
As a result, the choice of the correct 
filter would become a matter of the 
greatest complexity. The filter from the 
salmon-turquoise series which gave the 



Crandell, Freund and Moen: Illuminant Color Index 



393 











FILM IN 


DEX 


LIGHT 


SOJJKCE. 


xjgT 


^~ 

k 














4 15 16 17 1C 
BLUE - RED 



desired blue-red correction would also 
shift the green-red ratio. When allow- 
ance was made for the latter, and a 
suitable pink-green filter selected to 
supply the remaining green-red correc- 
tion needed, this filter would then upset 
the blue-red ratio! 

The final and truly logical solution 
which has been arrived at is best shown 
in Fig. 3A (same as Fig. 2, plus a grid 
representing the coordinates of the series 
of correction filters; 6 to raise color 
temperature, 6 to lower it, 3 to diminish 
green and 3 to increase it. Sliding the 
grid back and forth corresponds to the 
action of the computer.) As will be seen, 
the blue-red control filters (known as 
the CT or Color Temperature series) 
shift the illuminant along coordinates 
parallel to the black-body locus. The 
green-red control filters (known as the 
GC or Green Control series) shift the 
illuminant along coordinates perpendic- 
ular to the graph, which means that 
they alter the green-red ratio but do 
not significantly affect the blue-red 
figure. This releases us from the inter- 
dependence of the two filters involved 
in the previous proposal, and makes the 
second filter quite independent of the 
first. (See also Figs. 3B-3E.) 

A simple numerical example will in- 
dicate how this works out in p